Steel-Concrete Composite Bridge Design Guide September 2013

Steel-concrete composite bridge design
guide
September 2013
Raed El Sarraf, HERA, Auckland, New Zealand

David Iles, SCI, Ascot, United Kingdom
Amin Momtahan, AECOM, Auckland, New Zealand
David Easey, AECOM, Auckland, New Zealand
Stephen Hicks, HERA, Auckland, New Zealand
NZ Transport Agency research report 525
ISBN 978-0-478- 40769-3 (electronic)

ISSN 1173-3764 (electronic)
NZ Transport Agency
Private Bag 6995, Wellington 6141, New Zealand
Telephone 64 4 894 5400; facsimile 64 4 894 6100
research@nzta.govt.nz
www.nzta.govt.nz
The New Zealand Heavy Engineering Research Association (HERA) was contracted by NZTA in 2009 to carry
out this research
El Sarraf, R, D Iles, A Momtahan, D Easey and S Hicks (2013) Steel-concrete composite bridge design
guide. NZ Transport Agency research report 525. 252pp.
HERA, HERA House, 17-19 Gladding Place, Manukau 2241, Auckland, New Zealand, telephone 64 9 262 4756, facsimile
64 9 262 2856, structural@hera.org.nz, www.hera.org.nz
This publication is copyright NZ Transport Agency 2013. Material in it may be reproduced for personal
or in-house use without formal permission or charge, provided suitable acknowledgement is made to this
publication and the NZ Transport Agency as the source. Requests and enquiries about the reproduction of
material in this publication for any other purpose should be made to the Research Programme Manager,
Programmes, Funding and Assessment, National Office, NZ Transport Agency, Private Bag 6995,
Wellington 6141.
Keywords: bridges, composite, concrete, construction, cost-effective, design, durability, efficiency,

fabrication, fatigue, guide, ladder deck, life cycle, multi-girder, New Zealand, NZ Transport Agency, shear
studs, steel, Steel Bridge Development Group, sustainability, vibration, welding.
An important note for the reader

The NZ Transport Agency is a Crown entity established under the Land Transport Management Act 2003.
The objective of the Agency is to undertake its functions in a way that contributes to an efficient, effective
and safe land transport system in the public interest. Each year, the NZ Transport Agency funds innovative
and relevant research that contributes to this objective.
The views expressed in research reports are the outcomes of the independent research, and should not be
regarded as being the opinion or responsibility of the NZ Transport Agency. The material contained in the
reports should not be construed in any way as policy adopted by the NZ Transport Agency or indeed any
agency of the NZ Government. The reports may, however, be used by NZ Government agencies as a
reference in the development of policy.
While research reports are believed to be correct at the time of their preparation, the NZ Transport Agency
and agents involved in their preparation and publication do not accept any liability for use of the research.
People using the research, whether directly or indirectly, should apply and rely on their own skill and
judgement. They should not rely on the contents of the research reports in isolation from other sources of
advice and information. If necessary, they should seek appropriate legal or other expert advice.
Acknowledgements
This project was undertaken by the New Zealand Heavy Engineering Research Association (HERA), which is
based in Manukau, Auckland City, New Zealand. The researchers gratefully acknowledge the following
individuals and organisations for their support in the development of this document:
Dr Charles Clifton of the University of Auckland
Shane Culham of Culham Engineering
Jacques De Reuck of Metal Spray Suppliers (NZ) Ltd
Bob Delacey of George Grant Engineering
Alistair Fussell of Steel Construction New Zealand
Phil Gaby of Holmes Consulting Group
(SCNZ)
David Gifford of New Zealand Steel
Kenyon Graham of SKM
Matthew Holland of Tenix Robt Stone
Donald Kirkcaldie of Opus
Peter Lipscombe of URS
Jamie Macredie of Steltech
Willie Mandeno of Opus
Bruce Melsop of Eastbridge
Amin Momtahan of AECOM
David Moore of Grayson Engineering
Will Pank of BECA
Rob Presland of Holmes Consulting Group
Anand Reddy of Holmes Consulting Group
Craig Ross of Napier Sandblasting
Dr Wolfgang Scholz of HERA
Mike Sullivan of D&H
Tony Vos of Altex
Peter Wiles of Opus
This report is a key deliverable of a three-year project that was part-funded by the New Zealand
Transport Agency, the Steel Bridge Development Group and New Zealand Steel Industry representatives.
Abbreviations and acronyms

BSI
British Standards Institution
CAE
computer-aided engineering tools
CEN
European Committee for Standardization
FCM
fracture critical members
IZS
inorganic zinc silicate
NZTA
New Zealand Transport Agency
RRU
Road Research Unit, National Roads Board (predecessor of Transit NZ)
SH
state highway
SLS
serviceability limit state
SNZ
Standards New Zealand
ULS
ultimate limit state
Contents
Executive summary ...................................................................................................................... 7
Abstract ...................................................................................................................................... 8
1
Introduction ....................................................................................................................... 9
Typical composite bridge configurations ........................................................................... 10
2.1
Multi-girder bridges .................................................................................................. 10
2.2
Ladder deck bridges .................................................................................................. 14
2.3
Deck slab ................................................................................................................... 20
2.4
Shear connection ...................................................................................................... 21
2.5
Dealing with curvature and skew ............................................................................... 21
2.6
Substructures ............................................................................................................ 23
Benefits of steel-concrete composite construction ............................................................. 30

3.1
Why specify a composite bridge? ............................................................................... 30
3.2
Sustainability benefits ............................................................................................... 30
3.3
Economic benefits ..................................................................................................... 32
3.4
New Zealand case studies .......................................................................................... 34
3.5
Design concept ......................................................................................................... 37
Preliminary design .......................................................................................................... 38

4.1
General ..................................................................................................................... 38
4.2
Design for construction ............................................................................................. 38
4.3
Design for in-service maintenance ............................................................................. 42
4.4
Choice of structural configuration .............................................................................. 44
4.5
Preliminary sizing material selection ........................................................................ 45
4.6
Preliminary sizing multi-girder bridges ................................................................... 46
4.7
Preliminary sizing ladder decks ............................................................................... 48
Design standards ............................................................................................................ 49

5.1
The Bridge manual .................................................................................................... 49
5.2
AS 5100 .................................................................................................................... 50
5.3
NZS 3404 .................................................................................................................. 50
5.4
NZS 3101 .................................................................................................................. 51
5.5
BS 5400 .................................................................................................................... 51
5.6
Eurocodes ................................................................................................................. 51
5.7
Basis of design .......................................................................................................... 51
5.8
Project-specific requirements .................................................................................... 52
5.9
Health and safety regulations .................................................................................... 52
Calculation of action effects ............................................................................................ 53

6.1
Structural analysis ..................................................................................................... 53
6.2
Computer modelling ................................................................................................. 55
6.3
Deck slab analysis ..................................................................................................... 59
6.4
Design loading and load combinations ....................................................................... 60
6.5
Seismic effects .......................................................................................................... 65
Detailed design .............................................................................................................. 66

7.1
Overall bridge design ................................................................................................ 67
7.2
Main girders at ULS ................................................................................................... 69

5
7.3
Design for bending ................................................................................................... 71
7.4
Shear resistance ........................................................................................................ 81
7.5
SLS checks for compact and non-compact beams ...................................................... 82
7.6
Design of restraints to beams .................................................................................... 85
7.7
Longitudinal shear connection ................................................................................... 88
7.8
Connections .............................................................................................................. 94
7.9
Stiffeners .................................................................................................................. 99
7.10 Deck slab design ..................................................................................................... 104

7.11 Fatigue design ........................................................................................................ 108
7.12 Selection of steel sub-grade .................................................................................... 111
8
10
Construction considerations ........................................................................................... 113

8.1
Steel fabrication and concrete construction .............................................................. 113
8.2
Construction sequence ............................................................................................ 113
8.3
Cantilever edge slabs .............................................................................................. 114
8.4
Allowance for permanent deformations .................................................................... 114
Detailing considerations ................................................................................................. 116

9.1
Geometric configurations ........................................................................................ 116
9.2
Cross girder end connections .................................................................................. 119
9.3
Shear connection .................................................................................................... 123
9.4
Deck slab ................................................................................................................. 124
9.5
Bearing specification ............................................................................................... 126
9.6
Deck joints ............................................................................................................. 127
Design of durable composite structures .......................................................................... 128

10.1 Durability design of steel and concrete .................................................................... 128
10.2 An example of the durability design of steel ............................................................ 129
10.3 Coating application and inspection .......................................................................... 137
10.4 Maintenance management ....................................................................................... 138
11
Other types of composite construction ............................................................................ 139

11.1 Composite box girders ............................................................................................ 139
11.2 Network arch bridges .............................................................................................. 140
11.3 Stainless steel in bridge construction ....................................................................... 141
12
References .................................................................................................................... 145
Appendix A: Two-lane, three and four girder bridge costs .......................................................... 149
Appendix B: Available plate dimensions .................................................................................... 154
Appendix C: Initial sizing of main girders .................................................................................. 156
Appendix D: Background to the fatigue design criteria for bridges ............................................... 157
Appendix E: Single span ladder deck bridge .............................................................................. 160
Appendix F: Two span multi-girder bridge ................................................................................ 186
Appendix G: Single span multi-girder bridge ............................................................................. 231
Executive summary
Puentes mixtos de acero y hormign proporcionan una forma eficiente y rentable de construccin del
puente. Mediante la utilizacin de la fuerza extensible de acero de la viga principal y la resistencia a la
compresin del concreto en la losa, se incrementa la resistencia de flexin de los materiales combinados y
palmos ms grandes se hacen posibles.
En este documento se consideran dos tipos de compuestos puente. El tpica multi de la viga de acero y
hormign compuesta puente, que consiste en una serie de vigas de acero con refuerzo en el medio y una
losa en la parte superior y un puente escalera cubierta, que consta de dos vigas principales con un
nmero de secundarias vigas transversales intermedios que apoyan y actan con una losa de cubierta.
Ambos proporcionan una solucin rentable y la eleccin entre los dos tipos depende de consideraciones
econmicas y site-specific factores tales como la forma de los soportes intermedios y acceso a
construccin.
Este reporte de investigacin proporciona orientacin sobre las consideraciones generales para el
proceso de diseo preliminar y detallado, adems de orientacin sobre la verificacin de la adecuacin
estructural segn el manual del puente de agencia de transporte de Nueva Zelanda y el diseo
relevante y materiales estndares.
La gua describe la determinacin de las fuerzas de diseo, identifica caractersticas clave relacionadas
con el diseo de los diferentes componentes estructurales y da consejos detalle estructural. Tambin
proporciona orientacin adicional sobre filosofa y durabilidad rentable diseo.
El objetivo del documento es orientar a los principiantes y diseador puente experimentados en el diseo de
puentes mixtos de acero y hormign rentables.
Steel-concrete composite bridge design guide
Abstract
Este informe proporciona orientacin sobre el diseo de puentes mixtos de acero y hormign, que consisten en
vigas de acero y losas de hormign armado en la parte superior. Se consideran dos formas comunes: viga mltiples y
escalera cubierta puentes. Orientacin se da en las consideraciones generales para el proceso de diseo preliminar y
detallado, adems de orientacin sobre la verificacin de la adecuacin estructural conforme a los estndares
materiales y diseo NZ transporte Agencia puente manual y relevante. Tambin se proporciona orientacin adicional
sobre filosofa y durabilidad rentable diseo. El objetivo del informe es proporcionar orientacin para el principiante
y el diseador experimentado puente sobre el diseo de puentes mixtos de acero y hormign rentables.
Introduction
Introduction
Puentes mixtos de acero y hormign proporcionan una forma eficiente y rentable de construccin del
puente. Mediante la utilizacin de la fuerza extensible de acero de la viga principal y la resistencia a la
compresin del concreto en la losa, se incrementa la resistencia de flexin de los materiales combinados y
palmos ms grandes se hacen posibles. Este informe proporciona orientacin para el principiante y el
diseador experimentado puente sobre el diseo de puentes mixtos de acero y hormign, si el puente es
simplemente apoyados o continuo y de varias vigas y escalera cubierta formas de construccin. Tambin
se proporciona orientacin sobre el diseo de la filosofa y la durabilidad de diseo rentable. La gua
asume que el lector est familiarizado con los principios generales de diseo de estado lmite y tiene
conocimientos de acera estructural. Proporciona asesoramiento sobre las consideraciones generales para
el proceso de diseo preliminar y detallado, adems de orientacin sobre la verificacin de la adecuacin
estructural conforme a las instrucciones de puente...
Unless otherwise indicated, the guide refers throughout to the 3rd edition of the NZTA publication Bridge manual,
published in 2013.
Typical composite bridge configurations
El objetivo principal de un diseador de puente es proporcionar una solucin rentable de acuerdo con los
requerimientos del cliente. Puentes mixtos de acero y hormign de utilizan la fuerza extensible de acero
de la viga principal y la resistencia a la compresin del concreto en la losa para proporcionar una solucin
rentable en una amplia gama de vanos.
Los elementos de acero y hormign de un puente compuesto se conectan mediante conectores de
cortante que se sueldan a la brida superior de la viga de acero y estn integrados en el tablero de
hormign. La accin compuesta se logra mediante el esfuerzo cortante longitudinal transferido por los
conectores de corte, aumentando la resistencia de flexin significativamente en comparacin con el
que alcanz el rayo no compuesto.
Esta gua considera dos tipos de compuestos puente. El tpica multi de la viga de acero y hormign
compuesta puente, que consiste en una serie de vigas de acero con refuerzo en el medio y una losa en la
parte superior y un escalera cubierta puente, que consiste en medio de dos vigas principales con un
nmero de Cruz secundarios-vigas que apoyan y actan con una losa de cubierta. Ambos proporcionan
una solucin rentable y la eleccin entre los dos tipos depende de consideraciones econmicas y sitespecific factores tales como la forma de los soportes intermedios y acceso a construccin.
Otros tipos de construccin se resumen en el captulo 11, aunque ellos no ser cubiertos en detalle en este documento.
2.1
Multi-girder bridges
2.1.1
General
En la construccin del puente de vigas mltiples un nmero de similar tamaos vigas longitudinales se arregla en
espaciado uniforme en toda la anchura del puente, como se muestra en la seccin transversal tpica en la figura 2.1 para
una carretera de dos carriles con una barrera de trfico rgida y sin acera. La cubierta de la losa se extiende
transversalmente entre las vigas longitudinales y voladizos transversalmente fuera de las vigas del exteriores. Las vigas
estn arriostradas juntos en los soportes y en algunas posiciones intermedias. Accin compuesta entre la losa de cubierta
de hormign armado y las vigas longitudinales se logra mediante conectores de corte soldada con autgena en los
rebordes superiores de vigas de acero. Construccin de la viga mltiples puede ser usada para ambos solo se extiende y
abarca mltiples continua.
Figure 2.1
Tpico multi de la viga transversal de un puente compuesto con trfico rgidas barreras y sin acera
3%
3%
El arreglo que se muestra en la figura 2.1 es tpico cuando la losa se proyectara en encofrado temporal o prefabricados de hormign encofrado permanente de terrazas
se utilizan; muestra cuatro vigas de igual profundidad y con una superficie de la losa que sigue la inclinacin de la carretera. Una acera se proporciona a veces ambos
lados de la carretera, mientras que el trfico o barreras peatonales estn montadas sobre la losa de cubierta. La configuracin puede ser adaptada
10
para arreglos alternativos calzada, dados como en la Figura A1 del manual del puente. Diversos arreglos
para lidiar con comba y sper elevacin se discuten en la seccin 9.1.1.
2.1.2
Longitudinal girders
Vigas de acero son porta-paneles generalmente fabricadas-seccin; para pequeas luces, es posible
utilizar caliente laminados en vigas de seccin (vigas Universal) pero, por los motivos expuestos a
continuacin, laminado en secciones raramente son utilizadas en la construccin actual.
Alternativamente, las vigas soldadas (vase abajo) pueden usarse en algunas circunstancias.
Generalmente, las vigas estn espaciadas entre unos 3m y 4m de distancia y, por lo tanto, para un
ordinario dos carriles sobre el puente, normalmente se proporcionan cuatro vigas. Este arreglo es
compatible con la losa de cubierta (vase seccin 2.3), que distribuye las cargas verticales de las ruedas
en el doblez transversal.
2.1.2.1
Hot rolled sections Universal Beams
Universales vigas estn calientes en las secciones que estn disponibles en Nueva Zelanda hasta 612 mm
de profundidad como tabulados en tablas de capacidad de diseo para acero estructural (ASI 1999).
Otros tipos de secciones de laminado en caliente, algunas de las cuales son hasta 1016mm de
profundidad, estn disponibles desde el extranjero; Sin embargo, en Nueva Zelanda, Vigas soldadas se
utilizan comnmente para tales tamaos de viga (vase comentario sobre vigas soldadas ms abajo).
Puentes que utilizan vigas Universal suelen ser puentes de acceso rural carga ligera o puentes peatonales
que pueden abarcar hasta 12m simplemente apoyado o hasta 16 m para tramos continuos. Fabricacin
mnima se requiere para vigas Universal, slo con la soldadura de los refuerzos y perforacin de los
agujeros del perno de ser necesario.
2.1.2.2
Plate girders
Porta-paneles se fabrican de chapa de acero con arreglo a los requisitos del diseador. Proporcionan el
diseador con la flexibilidad de la especificacin del reborde diferentes tamaos y espesores web en
diversas posiciones a lo largo del tramo, para optimizar la viga, dependiendo de la duracin y aplicado en
el puente de carga. Sin embargo, el diseador debe ser consciente de la disponibilidad de tamaos de
placa y soldadura de requisitos, como seleccin inadecuado puede resultar en una costosa fabricacin;
optimizacin adecuada debe ser considerado a travs del proceso de diseo.
La profundidad de la viga placa inventados tambin puede variar a lo largo del tramo, segn sea necesario.
Es comn para aumentar la profundidad de la viga sobre el soporte intermedio donde se requiere capacidad
de momento negativo creciente. Para palmos por debajo de 50m, la seleccin de profundidad constante o
variado es a menudo dependiente de esttica. Para luces mayores de 50m, una profundidad variada puede
proporcionar un cierto grado de ahorro de costes en las regiones de medio palmo. La variacin en la
profundidad se logra mediante haunching recta (vigas cnicas) o curvar la brida inferior hacia arriba.
Longitudes de fabricacin de 26m o ms requieren el uso de las conexiones, debido a la limitacin del
transporte y fabricacin (ver apartado 4.2.1 para ms detalles). El nmero de conexiones depende de la
longitud requerida de la viga. Los empalmes de la viga son caros, atornillada o soldada. Excepto puentes
palmo largo, la solucin ms econmica es generalmente especificar los empalmes en o cerca del punto
de contraflexure.
2.1.2.3
Welded beams
Vigas soldadas son una forma de viga placa fabricada con dimensiones estndar (ASI 1999), que estn disponibles de 700mm a 1200mm
de profundidad. Pueden abarcar hasta 30m simplemente apoyado y hasta a 35m para tramos continuos. En los vanos, vigas se requiera ser
curvada de elevacin para el perfil de la carretera y pre convexo para carga muerta; Esto puede llevarse a cabo por los fabricantes de
puente utilizando maquinaria rodante para las vigas menos profundas pero agregar al costo. En muchos casos, Vigas soldadas podrn
sustituirse ms econmicamente por porta-paneles de profundidad similar pero en el que la brida tamaos y web han sido optimizados
grueso. Fabricantes pueden aconsejarle sobre la economa relativa.
11
2.1.3
Bracing
2.1.3.1
Support bracing
Vigas de acero deben tener un soporte juntos en las posiciones de apoyo, para la estabilidad y para
ayudar en el traslado de horizontal cargas (viento y fuerzas ssmicas) a los cojinetes o cizalla claves, las
cuales proporcionan transversal refrena en cada posicin de apoyo. El apuntalamiento de apoyo
generalmente es proporcionado por trianguladas apoyar sistemas, tales como un K-brace o una X-brace,
usando secciones de ngulo, o por las vigas horizontales (generalmente secciones del canal). Los
sistemas de arriostramiento en los soportes de extremos de puentes no integral pueden requerirse para
apoyar el extremo de la losa de cubierta. Puentes integrales requerir refuerzo en los soportes de
extremos para la condicin de construccin.
Un arreglo tpico refuerzo (utilizando K-apoyar) en un soporte intermedio se muestra en la figura 2.2.
Figure 2.2
Tpico refuerzo arreglo en un soporte intermedio, conocido como un K-brace (mostrado por una carretera
elevada super)
2.1.3.2
Intermediate bracing
Para puentes terminados, refuerzo intermedio es generalmente necesario en posiciones discretas en los
vanos de multi-puentes, para estabilizar los rebordes de fondo adyacente a los soportes intermedios
(donde estn en compresin). Durante la construccin, es necesario apoyar para estabilizar tanto las
pestaas inferiores adyacentes a los soportes intermedios y los rebordes superiores en las regiones de
medio palmo. Donde las vigas estn curvadas en plan, tambin se necesitarn refuerzos para proporcionar
contencin 'radial' a las pestaas inferiores (ver apartado 2.5.1).
En la mayora de los casos, el sistema de refuerzo ms eficaz es una estructura triangular entre vigas
adyacentes. En el puente ya terminado; Esto proporciona una trayectoria muy rgidas restricciones desde el
plano de la losa de cubierta a travs a las pestaas inferiores. En la etapa de construccin, refuerzo
intermedio entre vigas, sin apoyar plan, proporciona "restriccin torsional" (vase el debate sobre la
efectividad de tales arriostramientos en seccin 7.3.6). Como alternativa, 'refuerzo del canal' se utiliza a
menudo con vigas principales superficiales; el canal rgido est conectado rgidamente a las vigas
principales.
Refuerzo intermedio que es continua a travs de ms de dos vigas principales participar en la accin
global y va a distribuir la carga en cualquier uno carril a varias vigas principales. Sin embargo, dicha
continuidad no proporcionan mucho beneficio para el diseo de las vigas principales (especialmente
cuando el caso diseo considera todos los carriles cargados) e introduce las revocaciones de estrs en el
apuntalamiento y sus conexiones; los datos de conexin son potencialmente propensos a la fatiga. Para
evitar esta situacin de fatiga, diseadores pueden usar refuerzo no continua, donde las vigas principales
estn conectadas en pares, con ningn refuerzo entre un par y el otro, como se muestra en la figura 2.3.
Refuerzo intermedio tambin puede ser necesario si el espacio libre debajo del puente es tal que
la colisin de carga en el sofito puente debe ser considerado. Refuerzo en intervalos brinda
contencin a la brida inferior
12
2
y una ruta de carga a la cubierta del puente. En tales casos, el refuerzo en las ayudas tiene que ser
diseado para transferir la carga de colisin hasta los cojinetes o distorsionar las llaves. Orientacin sobre
las cargas de choque se da en la seccin 3.4.18 del manual del puente.
Figure 2.3
Typical paired bracing arrangements, X-brace (top) and the channel brace (bottom)
Aunque continuidad de arriostramiento transversal no es necesaria (y no siempre deseable, por la razn

mencionada anteriormente), puntal y atar los miembros son proporcionados a veces entre los pares de
rayos durante la construccin en orden a compartir las cargas de viento o para controlar el espacio entre
los pares. Tales miembros deba eliminarse una vez que la losa ha sido arrojada, debido a su participacin
estructural no deseado bajo trfico de carga. La remocin es una actividad potencialmente peligrosa que
debe considerarse cuidadosamente la hora de planificar el mtodo de construccin. Las conexiones al
refuerzo de construccin que queda en su lugar deben ser evaluadas para la fatiga de los miembros de
apoyo.
2.1.3.3
Plan bracing
Plan de refuerzo a la brida superior es una alternativa para proporcionar una rgida restriccin lateral a los
rebordes superiores en la etapa de acero desnuda durante la construccin. Aunque tal refuerzo es muy
eficaz en la brida de compresin en medio palmo de restriccin, su presencia complica la construccin.
Las dos posibles ubicaciones de refuerzo del plan estn por encima de la pestaa superior (conectada a
las calas en la pestaa superior) y por debajo de la brida superior. La parte superior de la brida superior
de refuerzo, no obstante, aade dificultad a la colocacin de refuerzo y los conflictos con el uso de
encofrado permanente como cubiertas prefabricadas. Refuerzo justo debajo de la pestaa superior puede
chocar con encofrado temporal y tendra que eliminarse despus de fundicin (ya que puede atraer las
fuerzas no deseadas cuando la losa se somete a carga local por encima de ella); tal refuerzo no es
recomendable y raramente se utiliza ahora internacionalmente.
Plan de refuerzo de vez en cuando se suministra a los rebordes de fondo de puentes estrechos cuando
los vanos son largas (ms de aprox. 60m) con el fin de mejorar la rigidez torsional total del puente (en la
etapa terminada) y as reducir la susceptibilidad a la inestabilidad aerodinmica. Esas mejoras en la
rigidez torsional tambin sera beneficiosa para un puente con curvatura significativa en plan. La
presencia de los refuerzos efectivamente crea un cuadro de pseudo.
2.1.4
Crosshead girders
A veces es conveniente reducir el nmero de columnas y los cojinetes de los soportes intermedios (ver seccin 3.5 para ms detalles). Por lo general, en lugar de un
cojinete directamente debajo de cada viga, un rodamiento es proporcionado a mitad de camino entre cada par de vigas, con una viga de cruceta para transferir las
reacciones. Tal arreglo es particularmente conveniente con grandes inclinaciones (ver seccin 2.5.2). Es un ejemplo de una cruceta
13
se muestra en la figura 2.4, que tambin muestra una viga de continuidad entre las vigas centrales; tan una viga es
ventajosa para la construccin, para reducir al mnimo giro durante el hormigonado, pero normalmente no es necesaria
para la condicin de permanente.
Figure 2.4
Crosshead girder in a multi-girder bridge
2.2
Ladder deck bridges
2.2.1
General
Escalera cubierta puentes son un acuerdo comn en el Reino Unido y tambin se estn volviendo
populares en Nueva Zelanda. Este tipo de puente proporciona slo dos vigas principales, con la losa
apoyada en cruz-vigas que abarcan transversalmente entre las dos vigas principales; la losa se extiende
longitudinalmente entre las vigas transversales. Este arreglo se denomina construccin 'cubierta de la
escalera', debido a la configuracin del plan de la acera, que se asemeja a los largueros y los peldaos de
una escalera.
Una seccin tpica de un puente de escalera cubierta se muestra en las figuras 2.5 y 2.6 (que ilustran la
configuracin 'escalera'). Cabe sealar que se construy el puente de cubierta de escalera en la figura 2.6,
con un pilar completamente integral para que las vigas principales y losa de cubierta eventualmente
fueron arrojados en un pilar rayo/diafragma en los extremos del puente, vea la seccin 2.6.4 y Iles (2010)
para obtener ms informacin. Adems, el refuerzo del plan se muestra es permanente pero slo fue
necesaria durante la construccin. Es ms rentable dejar el refuerzo que to quitarlo despus de la
terminacin de la construccin. Sin embargo, mantenimiento futuro y su costo deben considerarse como
parte del rgimen general de mantenimiento de la estructura. Para comentar sobre el uso de Vigas
voladizas ver seccin 2.2.4.
Figure 2.5
Cross section of a typical ladder deck bridge with rigid traffic barriers and no footway
3%
3%
14
El arreglo con dos vigas principales es apropiada (y econmica) para una anchura del puente hasta que
por una doble calzada de dos carriles. La economa viene principalmente de eliminar la cabeza viga
transversal y apoyar directamente a cada larguero en su propio conjunto de columnas, como se muestra
en la figura 2.6.
Las vigas principales y vigas transversales cuentan con conectores de cortante, para permitir la accin
compuesta a desarrollarse con la losa. Vigas transversales generalmente estn conectadas a las vigas
principales empernando; los refuerzos transversales intermedios son proporcionados en cada conexin
viga transversal.
Ms puentes de escalera cubierta estn diseadas con vigas principales de profundidad uniforme pero
pueden utilizar vigas de canto variable. En la figura 2.7 se muestra un ejemplo de una cubierta de la
escalera de haunched de la viga.
Figure 2.6
Steelwork arrangement of a curved ladder deck bridge (image courtesy of Holmes Consulting
Group)
Figure 2.7
Ladder deck bridge with haunched main girders (image courtesy of Holmes Consulting Group)
Donde la cubierta es amplia (ms de unos 22m), por ejemplo cuando se lleva una doble calzada de tres carriles, pueden utilizarse
dos arreglos de cubierta escalera adyacente. En tales casos, la losa de cubierta puede ser continua a travs de todas las principales
cuatro vigas o losas separadas pueden proporcionarse, uno en cada par de vigas. Cuando la losa es continua, atraviesa
transversalmente entre las vigas ms profundas (que as se limitan
a una distancia mxima de 3,5 m entre ellos). Donde se proporcionan las losas separadas, cada cubierta
voladizos transversalmente y se requiera algn tipo de articulacin en la reserva central; orientacin
puede encontrarse de los requisitos relativos a las brechas entre las cubiertas en NZS 3101 (SNZ 2006) y
TD19/06 (Agencia de carreteras 2006).
La losa de cubierta es continua, dependiendo de los arreglos de las calles y las barreras, momentos altos
pueden generarse en la cubierta. Esto puede requerir el uso de una viga transversal intermedia para hacer
la fila de vigas cruzadas continua a travs del ancho total del puente.
2.2.2
Main girders
Las vigas longitudinales principales son generalmente fabricadas porta-paneles; el ms pesado caliente
laminados en secciones (vigas Universal) o Vigas soldadas con autgena estn poco probable que sea
suficiente, incluso para palmos modestos. Porque hay slo dos webs, la placa de la web es ms gruesa de
lo que sera en un arreglo de viga mltiples; la esbeltez de web es menor y es generalmente posible
desarrollar la resistencia necesaria del esquileo en las webs sin uso de endurecimiento de la web, aparte
de eso en las vigas transversales.
Con los palmos largo, el tamao de las pestaas, particularmente la brida inferior, es probable que sea
bastante grande (en ancho y espesor). Es prudente para los diseadores para comprobar la disponibilidad
de material de placa adecuado en una etapa temprana, con especial atencin al grado de dureza y
dimensiones disponibles, como se discute en la seccin 4.5.1.
Como escalera cubierta puentes tienen slo dos vigas principales, plante la cuestin de redundancia
estructural podra estar en la opcin de configuracin de escalera cubierta (si algn evento accidental
fuera a daar una viga tan gravemente que ya no podra llevar incluso las cargas muertos, colapsara el
puente). No hay datos sobre la probabilidad de eventos accidentales que podran causar esos daos, para
la cubierta de la escalera o vigas varios puentes, y por lo tanto no es posible hacer cualquier evaluacin
cuantitativa de la fiabilidad de cualquier tipo. Las secciones de la viga de puentes escalera cubierta son
generalmente ms grandes que los de las cubiertas de vigas mltiples y ellos tambin son refrenados en
espaciado de cerca por las vigas cruzadas; por lo tanto, los diseadores consideran esta configuracin
sea suficientemente robusto para satisfacer requerimientos de redundancia estructural.
2.2.3
Cross girders
Vigas transversales son tpicamente espaciadas en entre 3,5 m y 4m centros, para adaptarse a un espesor
de losa de aprox. 250mm (vase la seccin 2.3). Por un sencillo de dos o tres carriles del puente, donde las
vigas principales son 7m-10 m aparte, secciones de laminado en caliente (vigas Universal) pueden ser
suficientes para los propsitos estructurales, pero placa viga secciones son ms propensas a ser utilizado.
Donde hay una inclinacin a la superficie de la carretera (por ejemplo, con una calzada nica de dos
carriles, como se muestra en la figura 2.6) la brida superior de una viga placa puede seguir la Cruz cada,
permitiendo el uso de un espesor uniforme de la losa y superficie. La brida inferior normalmente sera
recta. Si laminado en seccin transversal se utilizaron vigas, tampoco las secciones tendra que ser convexo
(que agrega al costo de fabricacin), o la losa o superficie tendra que ser afilados en grueso para
proporcionar las cataratas (que complica la construccin de losas).
Donde hay Super-elevacin de la superficie de la carretera, una viga principal se arregla ms alto que el
otro y la profundidad de la viga transversal es generalmente constante.
Vigas transversales son generalmente unstiffened y sin soporte, pero largas vigas transversales pueden
requerir refuerzos para la condicin de construccin (por lo general, canal de arriostramiento entre pares
de vigas en su tramo medio).
2.2.3.1
Intermediate cross girders in sagging moment regions
Vigas transversales intermedias actan eficazmente como simplemente apoyadas vigas en llevar
la carga de la losa. Los momentos finales, debido a la interaccin con las vigas principales, son
muy pequeos en relacin con la fuerza de
16
las vigas transversales, que por lo tanto pueden ser haban diseado como simplemente apoyadas vigas.
Sin embargo, los momentos finales pueden ser lo suficientemente grandes para influir en el diseo de la
viga transversal conexin viga principal.
En la condicin compuesta, las vigas transversales en las regiones de momento la flacidez de las vigas
principales estn obligadas a proporcionar contencin lateral a los rebordes de fondo de la viga principal
solamente donde las vigas principales se curvan en plan o lateral donde las cargas de impacto del
vehculo en el sofito son ser resistidas. Las vigas cruzadas ofrecen sujecin mediante la accin de marco
en U.
Las vigas cruzadas tambin proporcionan contencin hacia fuera-de-plano de la losa en compresin; la
fuerza de las vigas transversales y la esbeltez de la losa, ambos deben ser considerados (vase ms
discusin en secciones 7.6 y 7.10.2).
Durante la construccin, las vigas cruzadas proporcionan torsional moderacin a las vigas principales,
tanto como restriccin al lateral pandeo torsional y curvadas vigas principales, resistiendo las juntado
fuerzas generadas por las fuerzas de oposicin 'radiales' en las bridas de tensin y compresin.
2.2.3.2
Intermediate cross girders in hogging moment regions
En las regiones de momento tosco de las vigas principales, adyacente a las ayudas internas, la Cruz
intermedia vigas estn obligadas a proporcionar contencin lateral a los rebordes de la parte inferior de
las vigas principales, que son en compresin. Esta restriccin se proporciona a travs de los 'invertidos
U-Marcos' formados por las vigas transversales y a la cual se colocan los refuerzos. Las conexiones entre
principales y vigas cruzadas por lo tanto necesitan transmitir momentos de contencin y el marco debe
ser rgida. Si las vigas cruzadas son menos de la mitad de la profundidad de la viga principal, rodilla
vigorizante o haunched vigas cruzadas se necesite, tanto para rigidizar la estructura y reducir momentos
que deben ser transmitidos a travs de las conexiones de la viga transversal Meno (vase la seccin
2.2.3.3. para obtener descripciones de tales arreglos de apoyo Cruz vigas).
2.2.3.3
Cross girders at internal supports (pier diaphragms)
En los apoyos internos de vanos continuos, las vigas transversales muy a menudo son ms profundas
que las vigas cruzadas intermedias, proporcionando un ms rgido y ms fuerte 'diafragma muelle', con
las conexiones atornilladas que puede transferir las fuerzas de moderacin ms grandes que se
producen en los apoyos (ver figura 2.8). La viga transversal no debe ser tan profunda como las vigas
principales, para evitar conflicto con y conexin directa a la pestaa inferior de la viga principal.
Figure 2.8
Cross girder at an intermediate support of a ladder deck bridge
3%
3%
Como alternativa al uso de una viga transversal profunda, rtesis de rodilla o una cruz haunched viga puede proporcionar, como se muestra en las figuras 2.9
y 2.10. Esto endurece el marco y reducir momentos que necesitan ser transmitidos a travs de las conexiones de la viga transversal Meno y pueden ser
ventajosa si se conectan maneras servicios o acceso a los plafones de las vigas transversales a lo largo de la longitud del puente. En la prctica,
17
rtesis de rodilla intermedio raramente se especifican; es ms econmico usar una viga transversal
profunda. Haunched vigas cruzadas son un detalle an ms caro y un fabricante debera ser consultado
antes de seleccionar esta opcin.
Figure 2.9
Knee bracing arrangement
3%
Figure 2.10
3%
Haunched cross girder at an intermediate support
3%
2.2.3.4
3%
End supports
Con la construccin no integral, se utilizan los diafragmas de apoyo similares a los de los soportes
intermedios. Proporcionan un apoyo efectivo para el final de la losa de cubierta y la Junta de dilatacin.
Donde los soportes de extremos oblicuo al eje del puente, los diafragmas pueden actuar como vigas
trimmer (ver seccin 2.5.2). Para la discusin de pilares integrales, consulte Seccin 2.6.4.
2.2.3.5
Integral crossheads at internal supports
A veces cuentan con apoyos 'interior' de las vigas principales, bajo los diafragmas de muelle, en lugar de
directamente debajo de las vigas principales como se muestra en la figura 2.11. Los diafragmas son entonces ms
sustanciales y se refieren a menudo como 'crucetas integrales'. Puede haber buenas razones para tal arreglo,
especialmente cuando es difcil proporcionar apoyo debajo de una de las vigas de un puente oblicuo, pero agrega
considerablemente el costo de fabricacin y montaje. Si las vigas principales son haunched, tal arreglo, sin apoyo
directo debajo de los elementos ms cargados, puede causar preocupacin visual a algunos observadores.
18
Figure 2.11
Example of an integral crosshead
Jacking Stiffeners
Tenga en cuenta que si las vigas principales del arreglo que se muestra en la figura 2.11 eran haunched,
se necesitaran refuerzos en ambos lados de las webs de la viga principal y las conexiones web/brida
tendra que ser diseado para la carga de traccin debido a los componentes verticales de las fuerzas en
la brida de la viga principal inclinado.
2.2.4
Cantilever girders
Para longitudes normales del voladizo de la cubierta fuera de las vigas principales (hasta unos 2 m), no
son necesarias las vigas voladizas. Esto es porque la losa se voladizo transversalmente, como lo hace con
mltiples cubiertas de viga (vase discusin adicional en la seccin 2.3).
Vigas voladizas acero permiten voladizos ms cubierta ser proporcionado, pero la principal razn para
considerarlos sera para evitar la necesidad de encofrado voladizo o soportar barreras de trfico 'rgida' si
es necesario. Con Vigas voladizas, encofrado, en forma de terrazas de concreto prefabricado, puede
utilizarse a travs del ancho completo de la cubierta. En particular, las vigas voladizas son necesarias
cuando el hormign se coloca longitudinalmente a las vigas principales.
La provisin de Vigas voladizas conduce a la necesidad de continuidad de momento con las vigas
transversales. Esto agrega significativamente al costo de fabricacin. Adems, es difcil de lograr buena
alineacin en las puntas de voladizos largos y esto tambin se suma a costo. Una seccin transversal de
una baraja de escalera con Vigas voladizas se muestra en la figura 2.12.
Figure 2.12
Ladder deck with cantilever girders
Alternativamente, si longitudinalmente (es decir, perpendicular a las vigas principales) coloca concreto se
utilizan losas; las vigas voladizas pueden ser diseadas sin la necesidad de continuidad de momento. Esto
puede hacerse mediante el uso de profundidad se encuentra cubierta prefabricada borde losas con
refuerzo que se extiende en la losa de cubierta principal, que se convierte en parte de la losa principal una
vez el relleno de hormign in situ. Esto permitir que la losa de borde que se comporte como un elemento
tradicional de voladizo de la viga principal. Esto se muestra en las figuras 2.6 y 2.13, donde las vigas
voladizas se utilizan durante la etapa de construccin solamente pero queda en el puente despus de la
terminacin como una caracterstica visual..
19
Figure 2.13
2.3
Ladder deck bridge with cantilever beams (image courtesy of SKM)
Deck slab
Para resistir los efectos de la carga combinada de locales y globales de flexin (en particular, para las
cubiertas de la escalera, la flexin global en acaparar las regiones de momento los resultados en las
fuerzas de resistencia dentro de la losa) se necesita un espesor de losa de cubierta tpica de 250mm (el
valor depende en parte de los requisitos para la cubierta de refuerzo vea la discusin en la seccin
7.10). El refuerzo de la losa es tpicamente 20 barras en 150mm centros superiores e inferiores. Este
espesor de losa (hasta 250mm) puede estar en voladizo hasta unos 2 m y llevar aplicado cargas (longitud
total, del eje de la viga principal que fuera del borde de la losa). Cuando la losa de cubierta sufre
Accidente trfico cargando a travs de la barrera, el diseo de la cubierta debe asegurarse que la falla se
limita a la barrera slo con ningn dao a las piezas de fijacin a la cubierta, la losa de cubierta y
estructura de soporte. Alternativamente, las vigas voladizo pueden utilizarse, como se describe en la
seccin 2.2.4. Orientacin sobre el diseo de barreras se da en el Apndice B del manual del puente.
Echando todo el hormign in situ, en encofrado temporal, es el mtodo tradicional de construccin de
losa. Sin embargo, los costos de proporcionar y retirar el encofrado temporal, junto con problemas de
salud y seguridad para estas actividades, ha dado lugar a una preferencia por el uso de encofrado
permanente ya sea en forma de terrazas prefabricadas o cubiertas de acero perfilado. Espesor total
unidades prefabricadas tambin se utilizan.
En los proyectos ms recientes, profundidad parcial Cubiertas prefabricadas ha demostrado para ser la
solucin ms rentable para puentes mixtos en comparacin con el Deck in situ y profundidad
prefabricados losas (ver seccin 3.3 y Apndice A para ms detalles). Las unidades de profundidad parcial
actan inicialmente como encofrado permanente, y entonces el espesor de la cubierta restante es vertido
in situ. Proyeccin de refuerzo en la cara superior asiste en asegurar que los componentes prefabricados e
in situ actan plenamente como una losa. Cuando las tarimas de prefabricados se coloca transversalmente
a travs del puente, tampoco se coloca entre las vigas o los bolsillos estn diseados para permitir que los
conectores del esquileo dependiendo del tipo de Cubiertas prefabricadas utilizado. Si la tarima se coloca
longitudinalmente con las vigas principales, los bolsillos no son necesarios, como las unidades se colocar
en el borde de los rebordes de la viga, claros de los conectores del esquileo.
20
Uso de acero perfilado terrazas como encofrado permanente ha demostrado ofrecer costos beneficios en
comparacin con otros tipos de sistemas de terrazas (como se muestra en la seccin 3.4.1). Sin embargo,
la durabilidad es la principal preocupacin con esta forma de terrazas, tanto en trminos de la fatiga y la
corrosin. En tales casos, el diseador del puente debe seguir los requisitos pertinentes de la Agencia de
transporte de Nueva Zelanda (DZHENESI) para el diseo de este tipo de encofrado..
2.4
Shear connection
Hay un nmero de maneras de proporcionar una conexin de cizalla entre vigas de acero y la losa de
cubierta, pero de lejos el ms comn es el conector de Perno cabeza. Este es un taco hacia que se suelda
con autgena a la brida superior usando una herramienta especial de soldadura semiautomtica que
suministra un pulso elctrico suficiente para fundir el extremo de la espiga en la brida. En Nueva Zelanda,
la tecnologa est disponible para ser capaz de soldar clavos en el sitio, sin importar las condiciones
climticas. Sin embargo, mayora montantes se sueldan con autgena en el taller de fabricacin durante la
fabricacin de las vigas, que es la opcin ms rentable y evita el riesgo de dao a las capas protectoras.
Hay otras formas de conectores de corte, tales como soldadura de barras de acero con aros, placas
perforadas, placas soldadas en forma de T y pequeos tramos de canal. Aunque estos conectores
proporcionan una mayor resistencia de corte longitudinal por unidad, segn lo demostrado por Vianna et
al (2008) para placas perforadas y T-en forma de placas, cada uno han de ser soldado con autgena
manualmente y por lo tanto son ms caros..
2.5
Dealing with curvature and skew
2.5.1
Curved decks
Donde la cubierta del puente se curva horizontal (para adaptarse a la alineacin de carretera) las vigas
debajo de la losa pueden ser rectos o curvos en plan. Puede utilizarse una serie de vigas rectas con
cambio angular en posiciones discretas a lo largo de la longitud para radios grandes (ms de 300m);
Normalmente estos cambios podran estar en posicin palmo aproximadamente y , donde los
empalmes estn dispuestos en los puntos de contraflexure. La desventaja de tal arreglo es que la
longitud del voladizo vara a lo largo del puente. Aspecto, debajo del puente, se debe considerar
cuidadosamente al elegir esta opcin.
Los avances en la modelizacin informtica y control de equipos para la fabricacin han permitido
fabricantes cortar curvas bridas de la placa y as proporcionar 'verdadero' curvado vigas. Esto supera el
problema de voladizos de longitud variable y proporciona una mejor apariencia debajo del puente. (En la
prctica, las placas brida todava se cortan como una serie de rectas, pero son tan corto, 1m o menos,
que parecen verdaderamente curvado).
El cambio de direccin de la brida inferior, en posiciones discretas o 'continua' requiere una 'radial' de la
fuerza para equilibrar el cambio en la direccin de la fuerza de la brida. Con vigas mltiples cubiertas,
arriostramiento transversal se requiere en las posiciones de cambio entre una serie de rectas o a intervalos
a lo largo de una viga curva (el intervalo necesitado depende de la curvatura y la anchura del reborde). Con
las cubiertas de la escalera, el espaciado regular de las vigas transversales y su apego a las vigas
principales es capaz de proporcionar esta contencin lateral a la brida; las vigas transversales son
dispuesto radial a la curva.
2.5.2
Skewed bridges
2.5.2.1
Para puentes de viga multi sesgadas, casi siempre se arregla refuerzo intermedio cuadrado a las vigas
principales; No hay ninguna ventaja particular en alinear tales refuerzos sobre el sesgo para pequeos
ngulos de inclinacin y, para grandes ngulos de inclinacin, la interaccin con el doblez de las
complicaciones de las vigas principales causas en el diseo.
21
En intermedio y final soportes, enganches se arreglan generalmente en la lnea de los soportes de

inclinacin para ngulos de inclinacin pequeos (menos de 25); para grandes ngulos de inclinacin,
refuerzo en los soportes intermedios es generalmente cuadrado a las vigas principales pero es refuerzo en
los extremos a lo largo de la lnea de los soportes. Arreglos vigorizante tpicos se muestran en la figura
2.14. Note que para pequeas inclinaciones crucetas integrales son generalmente continuos (aunque las
vigas de continuidad entre las vigas principales interiores son mucho ms ligeras que los sobre el apoyo).
Para inclinaciones mayores no son sin vigas de continuidad entre las vigas del interiores, para evitar
posibles problemas de fatiga, aunque arriostramiento continuidad puede ser necesaria para que la
construccin, para el control de giro en la etapa de hormign hmeda. Mayor orientacin en posicin
oblicua se encuentra en la seccin 4 del manual de puente con orientacin complementaria adicional en
direccin no.1.02 nota de Hendy y Iles (2009).
Figure 2.14
Arrangements for skewed multi-girder decks
Intermediate support
4 bearings
Support
bracing
Deck axis
2 bearings
Intermediate
bracing
Integral
crosshead
(a) Skew
< 25
2 bearings
4 bearings
Support
bracing
Intermediate
bracing
Integral
crosshead
Deck axis
2.5.2.2
(b) Skew > 25
Ladder deck bridges
Skewed intermediate supports
Un mrito particular del escalera cubierta acera sistema es que sesgada soportes intermedios pueden ser fcilmente
acomodados, como uno de los extremos de una cruz viga puede ser conectada a la pestaa de rodamiento sobre el
apoyo a una viga mientras que el otro extremo puede ser conectado a un refuerzo en el lapso intermedio. Con tal
disposicin no ser necesariamente en un espaciado regular a lo largo de la longitud de la cubierta de las vigas
transversales, pero se espaciarn dictadas por la geometra de la posicin oblicua (ver figura 2.15).
22
Figure 2.15
Arrangement at skewed intermediate support

CL Intermediate Support
Note: Only the girder webs and the web stiffeners are shown, for clarity.
Skewed end supports

En los soportes de extremo para sesgar, recortadas vigas cruzadas, conectadas a una viga trimmer final,
pueden ser necesarias, como se muestra en la figura 2.16. Este arreglo es generalmente preferible a un
arreglo 'aviv' de vigas cruzadas. Para simplificar los detalles de conexin, las conexiones en la esquina
obtusa para el trimmer final y la viga transversal estn separadas, aunque las consecuencias sobre el
diseo de losa en esta rea deben ser consideradas cuidadosamente y modelado 3D puede ser necesarias
para predecir el comportamiento local con suficiente exactitud.
This arrangement is used even with integral abutments; the trimmer beam is then cast into the endscreen wall.
Figure 2.16
2.6
Arrangement at skewed end support
Substructures
Subestructuras suelen ser construccin de hormign armado y consisten en los componentes estructurales de los
puentes que estn por debajo de la superestructura los pilares, pilares, muros de contencin, encepados y pilotes.
La superestructura puede expandirse y contraerse por el uso de juntas de expansin en el nivel de la calzada y
'descanse' sobre cojinetes estructurales sobre los pilares y soportes intermedios. Sin embargo, el uso de
23
columnas tubulares acero llenadas de hormign pueden ser rentables, con la orientacin de diseo disponible en NZS 3404 (SNZ
1997b; 2009).
La subestructura del puente es un componente importante del costo total del puente compuesto y es
donde la mayora de los ahorros de costos puede lograrse debido al peso ligero de puentes mixtos de
acero y hormign (demostrado en la seccin 3.4). Normalmente son 40% ms liviano en general en
comparacin con puentes de hormign; subestructuras de tal modo ms ligeros pueden ser utilizados. El
diseo de infraestructuras est fuera del alcance de esta publicacin; Sin embargo, se recomienda que el
diseador est familiarizado con los requisitos de diseo de la subestructura que figuran en los apartados
4 y 6 del manual de puente, as como el efecto de la subestructura en el costo total de la estructura.
A lo largo de los aos, experiencia ha demostrado que el uso de los resultados de las articulaciones y los
rodamientos de expansin en problemas relativamente costoso mantenimiento que requieren reemplazo y
mantenimiento regular. Esto es debido a desgaste normal y la penetracin de suciedad y escorrenta
contaminada camino que resulta en extenso deterioro de la estructura como un todo. Esta experiencia ha
llevado al desarrollo de puentes integrales, donde no hay ninguna articulacin a nivel de la calzada y la
estructura de soporte se ve obligada a desplazar con los movimientos de la cubierta. Diferentes formas de
construccin integral pilar que se utilizan estn resumidas en la seccin 2.6.4. Diseo detallado de pilares
integrales est fuera del alcance de esta publicacin, pero algunas de las cuestiones de detalle estn
cubiertos en Iles (2010) y reproducidos en la seccin 2.6.4.
En los puentes no integral, la cubierta se sienta en los cojinetes que son compatibles con los estribos y
pilares intermedios. Estribos pueden extenderse a zapatas o pueden ser apoyados sobre pilotes; los
estribos tambin pueden actuar como muros de contencin de altura completa. Soportes intermedios
pueden tomar la forma de columnas individuales (uno debajo de cada cojinete) o de una pared o 'hoja
muelle' apoya todos los cojinetes en esa posicin intermedia. Discusin sobre las formas de estas ayudas
est fuera del alcance de esta publicacin, pero los mecanismos de articulacin se discuten ms abajo.
2.6.1
Bridge articulation
En los puentes no integral, la cubierta del puente est soportada sobre rodamientos en cada soporte y
sujecin lateral se proporciona en cualquiera algunos de estos rodamientos o por cizalla claves
encuentran ambos lados de los cojinetes. Este arreglo de las restricciones, que permite la expansin
trmica y la contraccin de la cubierta, se conoce como articulacin. Una disposicin tpica para un puente
dos apoyado sobre cojinetes de marihuana se muestra en la figura 2.17; arreglos alternativos y una
discusin general de articulacin se dan en la nota de orientacin nm. 1.04 Hendy y Iles (2009).
Figure 2.17
Articulation of a two-span bridge, where fixity is provided by bearings
Symbol
Freedom
Translation fixed, rotation free.
Translation in one direction,
rotation free.
Translation free, rotation free.
Cuando la cubierta es curva en plan, la alineacin de rodamientos guiadas debe ser considerada
cuidadosamente, puesto que la cubierta intenta aumentar o disminuir en radio, as como
ampliar/contrato de longitud. Tambin se incluyen ejemplos de articulacin para las cubiertas curvadas
en nota orientativa N 1.04 Hendy y Iles (2009).
Para un puente totalmente integral, hay no hay libertades en los soportes del final pero todava hay una
opcin para hacerse de la libertad/moderacin en los soportes intermedios; una gua del cojinete se
proporciona generalmente en cada soporte intermedio.
24
2.6.2
Intermediate supports
2.6.2.1
Multi-girder decks
Vigas mltiples cubiertas son compatibles en las posiciones intermedias en muelles de hoja, bents
columna, columna circular solo vstago con cabezas de martillo o columnas individuales en cada viga.
Columnas individuales en cada viga pueden aparecer algo desordenadas en algunas situaciones y un
arreglo alternativo es poner una columna entre cada par de vigas y utilizar una cruceta continua integral
entre las vigas.
2.6.2.2
Ladder decks
Las cubiertas de la escalera son generalmente sostenidas por columnas directamente debajo de cada viga
principal; este modo consigue una apariencia ms bajo el puente, as como la economa. Para cruces de
ros, muelles de hoja o Tajamar toda su longitud (es decir un muelle largo puente con extremos en forma
de cua) con columnas de arriba puede ser preferido por razones de Hidrologa. Requisitos para la
sustitucin de los rodamientos pueden dictar el tamao mnimo de las columnas, como es preferible que
tomas pueden ser colocados en la parte superior de cada columna, permitiendo as que la acera a ser las
columnas para reemplazar los rodamientos de la paja.
Muelles de hoja, en lugar de columnas individuales, a veces se utilizan cuando los cojinetes del puente se
encuentran dentro de las vigas principales y reciben llamadas cabezas Cruz integrales (ver figura 2.11).
2.6.2.3
Intermediate supports for bridges with integral abutments
No hay ningn requisito para vigas hacerse integral con soportes intermedios cuando el puente est
diseado como un puente integral. Para ello aade complejidad con poca ventaja y deben evitarse. La
referencia a 'crucetas integrales' arriba no indican la construccin integral entre los sub - y
superestructura. Sin embargo, las vigas principales sern generalmente continuas por encima de los
soportes intermedios.
2.6.3
End supports - non-integral abutments
Para la construccin no integral, sern necesarios los cojinetes bajo las vigas principales y una junta de
expansin deber proporcionarse en la cubierta. En el Reino Unido, una galera de inspeccin tambin
se encuentra debajo de la Junta de dilatacin; Sin embargo, esto no es obligatorio en Nueva Zelanda,
con el costo de la instalacin de una galera de agregar que el costo de construccin, especialmente
para pequeos y medianas puentes. Aunque es una buena idea en principio, el diseador debe tener en
cuenta la implicacin del costo de tal una adicin a los costos totales de puente.
Un arreglo tpico de un pilar no integral para un puente de escalera cubierta, con golpe de detalle que se
requiere en el diseo ssmico, se muestra en la figura 2.18. Cabe sealar que las siguientes figuras Pilar
excluyan las losas de asentamiento para mayor claridad, que son requeridas por el manual del puente.
En cubierta escalera puentes donde las vigas principales son ampliamente espaciadas o el extremo de la
cubierta es muy sesgada, la desviacin vertical del extremo viga transversal entre dos rodamientos
podra ser mayor de lo que puede dar cabida a una junta de dilatacin (3mm es considerado el mximo
para las juntas utilizadas). Si este es el caso, debera proporcionarse cojinete intermedio uno o ms
debajo de la viga transversal. (Sin embargo, el caso econmico debe considerar cuidadosamente, ya que
puede ser menos costoso proporcionar material adicional en la viga o lo encajonar en concreto para
aumentar la rigidez.) Si se proporciona un cojinete intermedio, el rodamiento puede ser necesario
precargado para evitar chattering2 o que se refrenar contra levantamiento.
The dynamic effects of traffic loading on the deck may at times cause upward load effects on the end cross girder
and, if there is very little dead load on such a bearing, the end cross girder may deflect upward and lift off the bearing.
Lift off and subsequent impact on closing is often referred to as chattering. This behaviour is very onerous in terms
of bearing life and must be avoided.
25
Figure 2.18
Non-integral abutment with knock off detail

Expansion joint
Knock off detail
2.6.4
End supports - integral abutments
Para puentes hasta 55m de eslora, pilares integrales pueden utilizarse si el ngulo de sesgo no se trata ms
de 30 , como se indica en la seccin 4.7.1(c) del manual del puente. Ms puentes integrales son aceptables
proporcionando anlisis adecuados mtodos se utilizan. Existen tres tipos de construccin integral que
puede ser utilizado para puentes mixtos en Nueva Zelanda:
fully integral bridges framed abutments
fully integral bridges bank pad abutments
semi-integral bridges- with bearings
The forms of these abutments are discussed below; design and detailing issues are discussed in detail in
Iles (2010).
2.6.4.1
Framed abutments
Framed abutments are usually built with H-piles or reinforced concrete piles, with the piles inside sleeves,
thus avoiding earth pressures on the piles as the bridge expands and contracts (Iles 2010). A typical
arrangement is shown in figure 2.19, with a normal earth slope in front and in figure 2.20 with a
reinforced earth retaining wall. Depending on soil conditions, one or two piles are provided for each main
girder in multi-girder bridges; for a ladder deck bridge of the same overall width, a similar total would be
provided, though they might be concentrated around the positions of the main girders. Framed abutments
are also built with reinforced concrete abutment walls on strip footings, although that form of
construction is not discussed in this guide.
26
Figure 2.19
Framed integral abutment - with normal earth slope
End diaphragm
Endscreen
Temporary bearing
Pile cap
Pile (RC or H-pile)
Sleeve (if needed)
Figure 2.20
Framed abutment with reinforced earth retaining wall
End diaphragm
Temporary bearing
Pile cap
Reinforced earth
retaining wall
Pile (RC or H-pile)

Sleeve (if needed)
In principle, any type of bearing pile, including steel H-piles, can be driven into the ground and the endscreen wall cast around the tops of the piles. In practice, only a small number of bridges have been built
with H-piles. Where construction has used H-piles, they have usually been encased in a pile cap just below
the bottom of the main girders. Plates for temporary bearings are set into the pile cap and the end-screen
wall is completed later, after the deck steelwork has been erected and the deck slab cast.
With fully integral construction, bracing for the construction condition may be arranged within the wall
(and will be cast in) or just in front of it (but then there must then be access for maintenance).
2.6.4.2
Integral bank pad abutment
In an integral bank pad abutment, an end-screen wall is cast around the ends of the girders and this wall
sits directly on the soil beneath. A typical arrangement is shown in figure 2.21. Guidance on the type of
deck joint to use is given in section 4.7.4 of the Bridge manual. Finally, even though the porous block
work and drainage system is not currently used in New Zealand, designers should consider the benefits of
reducing water pressure build-up and reduction of potential durability issues in future bridge design.
27
Figure 2.21
Integral bank pad abutment

Joint
Porous blockwork and

drainage system
Cast-in
temporary
bearing
Revetment
Construction Joint
Because the expansion and contraction of the deck causes the foundation to slide and rotate on the soil,
the design bearing resistance of the soil has to be reduced. This type of abutment is better suited to
situations where the soil is non-cohesive (or where cohesive material has been dug out and replaced with
non-cohesive material).
2.6.4.3
Semi-integral abutment
In a semi-integral abutment an end screen wall is generally provided across the end of the deck, with the
girders supported on bearings in front of the wall. A typical arrangement is shown in figure 2.22. This
form of abutment can be used either with revetment slopes in front of the abutment or behind a retaining
wall. It is particularly suitable where there is a reinforced earth retaining wall. However, replacement of the
bearings will require jacking and because of concerns about the forces involved and the movement at the
interface with the soil, it is a less favoured solution (Iles 2010).
Figure 2.22
Semi-integral abutment
Porous blockwork and

drainage system
End Screen
Construction Joint
Sliding Bearing
Footing
A semi-integral abutment is only suitable for up to about 15 skew because with larger skews the lateral
component of earth pressure exerts large transverse forces on the bearings.
28
With semi-integral construction, the end-screen wall is usually connected to endplates across the ends of
the girders. The end-screen wall will act as torsional restraint to the girders and as a trimmer beam. Some
form of restraint to the main girders, either within the wall or in front of it, will be required for the
construction condition.
With wide ladder decks, there is potentially a similar concern about excessive vertical deflection of the
end-screen wall as noted above for the end cross girders in non-integral bridges but usually the wall is
sufficiently stiff for deflections to be small.
29
Benefits of steel-concrete composite

construction
3.1
Why specify a composite bridge?
Corta la mayora abarcan puentes en Nueva Zelandia (y en muchos pases alrededor del mundo) con
vanos hasta 30m se han construido en concreto, normalmente compuesto por vigas de hormign
prefabricadas y cubiertas de hormign. Sin embargo, un puente compuesto de acero y hormign bien
diseado formada por vigas de acero y hormign cubiertas tambin proporcionar una solucin
econmica y sostenible. Figura 3.1 muestra que el potencial se extiende al tipo de superestructura para
ambos puentes de acero y hormign compuestos y concretas, aunque se extiende ms puede ser posible
que algunas de las opciones mostradas.
Aunque puentes de acero se utilizan comnmente en medianos palmos (entre 30m a 80m) y
especialmente para el diseo, la orientacin y palmos largos (ms de 80 m) filosofa descrita en esta
seccin proporciona el diseador con las herramientas necesarias para disear un compuesto de acero y
hormign rentable puente, si es un corto, mediano o largo palmo estructura.
Figure 3.1
Span to superstructure type for road bridges
Box Girders
Concrete Composite
Trough Girders
Plate Girders
Welded Beams
Hot Rolled Beams
Box Girders
Super T
Hollowcore
I Beam
5m
10m 15m
20m 25m 30m 35m 40m 45m 50m 55m 60m 65m 70m 75m 80m
Economic Span Range
3.2
Uneconomic Span Range
Sustainability benefits
La combinacin de acero y concreto en un nico elemento estructural compuesto mejora las ventajas
individuales de ambos materiales. Utilizando la fuerza de alta resistencia de acero junto con la resistencia a
la compresin del hormign, los elementos resultantes tienen una y media o incluso doble la fuerza y la
rigidez en comparacin con un elemento no compuesto. Esta es la principal ventaja de la construccin del
puente compuesto, que es reconocida en todo el mundo y la razn de por qu es generalizada en el Reino
Unido, Europa y Amrica del norte.
En lo que respecta a los aspectos de sostenibilidad de ambos materiales, hay una serie de artculos y
documentos escritos sobre este tema, tales como la construccin sostenible de acero (Corus 2006) y
Concrete3 econmico, social y ambiental (CCANZ 2007). Esta seccin proporciona una visin general de los
beneficios de la sostenibilidad de elementos de acero, hormign y compuestos.
30
Benefits of steel-concrete composite construction
3.2.1
Supply
Acero y el hormign estn disponibles en Nueva Zelandia y desde el extranjero.
3.2.2
Recycling and reuse
Un atributo importante sostenibilidad del acero es su capacidad para repetidamente ser reutilizado o
reciclado sin cualquier degradacin en las propiedades del material de calidad y mecanizadas. Otros
materiales son a menudo reciclados solamente una vez antes en ciclismo, que significa finalmente
encontraron su camino a vertederos.
Acero nunca pierde su valor y tiene un ciclo de vida econmico sostenible que no tiene rival por la
mayora de los otros materiales de construccin. Todo el acero usado tiene un valor, si es ser reutilizado
o reciclado. Esta propiedad reciclaje ya estaba siendo utilizada antes de sostenibilidad se convirti en un
problema. No haba ninguna necesidad de legislar para que ser reciclados, como el acero tiene un valor
intrnseco como material de desecho y est siempre en demanda de la produccin de acero nuevo.
En lo que respecta al concreto, debido a la limitada oferta local de agregado en algunas zonas alrededor de
Nueva Zelanda (como Auckland), es ahora una opcin viable para concreto agregado en instalaciones
especialmente diseadas para reciclar. Estas plantas son capaces de aceptar, procesar y vender hormign
reciclado agregado. rido reciclado de hormign se separa de su refuerzo acero reciclable y se procesa en
tamaos especficos agregados.
3.2.3
Waste minimisation
Poco material de desecho que se genera durante la fabricacin de componentes de acero, y la mayora de
esto es recuperado y reciclado. En sitios de construccin, que a menudo pueden generar grandes
volmenes de residuos de otros materiales, fabricacin inspecciones asegura que no se desperdicie acero,
como slo lo que se necesita viene a sitio. Casi todos los materiales residuos generados en el taller de
fabricacin es recuperado para reutilizacin o reciclaje.
Con mayor entendimiento del impacto del agua y eliminacin de residuos, ha habido progresos
considerables en la reduccin del vertido de aguas residuales a travs de las instalaciones de produccin
de la industria concreta. Ahora es una prctica comn para utilizar sistemas de lavado qumico de
residuos y recuperadores agregadas para minimizar lavar desechos y agua de la limpieza de los tazones
de fuente mezclador del carro y planta.
3.2.4
Off-site manufacture
Fabricacin fuera del sitio ha sido siempre una caracterstica clave de la construccin de acero como es
para hormign prefabricado construccin como puente Deck, que permite una estructura compuesta de
puntuacin alta en muchos criterios de sostenibilidad.
Los componentes ms precisos pueden lograrse con el compuesto de elementos fabricados y fabricadas
fuera de sitio. Residuos se minimizan y productos de alta calidad, libres de defectos son posibles. En los
talleres de fabricacin moderna y yardas de fundicin previa, donde la maquinaria de ltima generacin
de control numrico est completamente integrada con diseo asistido por ordenador (CAD) y otro
software, elementos compuestos pueden ser fcilmente estandarizados, probados y certificados.
Recubrimientos de proteccin a la corrosin pueden aplicarse a elementos de acero en la etapa de
fabricacin, reducir el programa general de construccin del sitio. Recubrimientos anti-grafitis tambin
pueden aplicarse a elementos de hormign si es necesario.
Las comunidades locales se benefician de fabricacin fuera de sitio, ya que hay menos trfico a sitios
dando como resultado menos congestin de trfico local. Otro beneficio para la poblacin local es que la
construccin prefabricada es seca, polvo - libre y relativamente tranquilo.
Elementos fabricados fuera del plomo a los programas de construccin ms predecibles. Los administradores de sitios se benefician de just-in-time entrega,
siendo capaces de mantener estos elementos en depsitos o en el taller de los contratistas
31
hasta que se necesite, ahorrando espacio y reducir la posibilidad de daos de almacenamiento in situ.
Una vez entregada al sitio elementos prefabricados son rpidamente y con seguridad erigidos. Elementos
prefabricados tambin son inherentemente ms seguros, que requieren generalmente bien ensen $ ados
y menos personas instalarlos. Las actividades del sitio son predecibles y bien practicado.
3.3
Economic benefits
Apndice A resume los hallazgos de un estudio de comparacin de costos para tres y cuatro vigas de
opciones para un puente de dos carriles; Esta es una versin abreviada de El Sarraf (2008). Los resultados
de los puentes con los tramos de 12m, 15m, 21m, 27m y 30m, ha sido reconfigurado como una
representacin grfica para la facilidad de ver en la figura 3.2. El eje y representa, en el siguiente orden, el
puente del palmo, nmero de vigas, nmero de apoyos y el tipo de cubierta (que se da en la tabla A.2 en
apndice A6). El eje x representa el costo del puente menos la tarima y luego por separado las tarimas de
costo. La suma de estas cifras proporciona el costo total del puente.
En la tabla A.4 y figura 3.2, la diferencia porcentual entre las diferentes opciones del decking indica que la
profundidad parcial decking prefabricado es la opcin ms barata, seguido por la tarima in situ y luego
seguido de cerca por la profundidad prefabricado de terrazas. Esto es consistente con la regeneracin de
contratistas. La otra cuestin que determina el uso del tipo de cubierta es el refuerzo doblez requisitos
(vase la seccin 4.2.3).
Figura 3.2 muestra tambin que los elegidos decking inventa un gran porcentaje del costo total del
puente. Costo del decking concreto puede explicar hasta un 55% del costo total de los puentes palmo ms
cortos pero puede ser tan bajo como el 21% del costo total de puentes de tramo ms largo. A medida que
aumenta la duracin, el cociente de las tarimas costo total costo disminuye, haciendo la eleccin correcta
de las tarimas especialmente crtico para puentes palmo ms cortos. Un resultado interesante es que la
opcin de cuatro-viga proporciona una solucin ms rentable que la opcin de tres vigas; Esto demuestra
que el principal problema que los diseadores deben ser conscientes de que el costo de tonelaje de acera
total es el factor que gobierna y no el nmero de vigas utilizado (este tonelaje incluye el peso de las
unidades de refuerzo).
Tambin es significativo el efecto del nmero de unidades de refuerzo en el tamao de la viga principal.
Frenos son necesarios durante la ereccin de la superestructura para proporcionar contencin lateral a la
brida superior hasta que la cubierta se coloca y actos compositely con las vigas principales; el menos
apuntalamiento utilizado, el ms grande del tamao de la viga principal.
Teniendo en cuenta todos estos factores, los resultados sugieren que la ptima opcin para un puente de
acero u hormign compuesto de terrazas es la profundidad parcial prefabricada con el menor tonelaje
acera independientemente del nmero de vigas.
32
Figure 3.2
Graphical representation of the two-lane, three and four-girder bridge costing data
33
3.4
New Zealand case studies
The following bridge case studies provide a cost comparison between a conventional concrete bridge and
steel-concrete composite ladder deck system (see section 2.2).
3.4.1
Weiti Stream Bridge
A finales de 2006, un puente privado tres fue diseado originalmente con una cubierta convencional
doble ncleo hueco sobre vigas de cruceta de muelle. El lapso de 17m est dentro del rango ptimo para
las unidades de ncleo hueco. Sin embargo, cambiando la superestructura de un sistema de cubierta de
acero u hormign compuesto escalera, tomando ventaja de la continuidad durante los tres palmos y
eliminando las costosas vigas de cruceta, se obtuvo una solucin ms rentable. El escalera cubierta
sistema consisti en dos vigas principales en los centros de 9m, con 18 intermedios cruzan las vigas en
los centros de 3 m. Inicialmente, los elegidos decking sistema para el puente compuesto eran que una
profundidad parcial prefabricados de hormign cubiertas; Sin embargo, debido a cuestiones que afect a
la fecha de entrega del proyecto de fabricacin, una opcin de plataforma de acero era considerado como
encofrado permanente. Los elegidos decking sistema fue entablado un Tata internacional ComFlor 80
(Tata Internacional 2005) acero con un total de 330 mm in situ concreto rellenado de para disminuir las
necesidades de encofrado. La plataforma w....
Es claro que las opciones de puente de acero compuesto eran ms rentables que el ncleo hueco
convencional hormign decking, una vez que el costo total del proyecto fue considerado de tabla 3.1. La
superestructura de los costos de las tres opciones son comparables, con la mayor parte de los ahorros de
costo atribuidos a la subestructura, especialmente con el retiro del muelle Cruz cabeza rayo y la
sustitucin de los estribos con un muro de contencin de concreto segmentario. Un dato interesante es
que la opcin de superficie de acero proporciona mayores ahorros de costos que la profundidad parcial
prefabricados hormign opcin terraza, con un ahorro de NZ$ 37.492 entre las opciones de dos puentes
de acero compuesto. El puente de acero y hormign compuesto proporcion un total de ahorros de NZ$
76.295 en comparacin con el puente de hormign doble ncleo hueco.
Se lograron ahorros ms quitando el estribo y reemplazarlo con un muro de contencin de concreto
segmentario con un relleno detrs de la pared. El relleno haba conducido pilotes de madera para apoyar la
gra teniendo los cojines durante la construccin.
Este puente fue para un cliente privado y tena una vida de diseo especificada de 50 aos, mientras que
el manual del puente requiere 100 aos. Aspecto no era una preocupacin; por lo tanto acero
desprotegida fue utilizado con un subsidio de diseo para la prdida de acero contra la corrosin de
acuerdo con El Sarraf y Clifton (2011). Sin embargo, el costo de un sistema de recubrimientos para dar un
perodo de 35 aos antes de su primer mantenimiento habra sido NZ$ 28.000, dando un ahorro neto
para un puente cubierto de NZ$ 48.295 para la opcin de puente de acero compuesto de superficie de
acero.
Cabe sealar que existen limitaciones en el uso del acero desde una perspectiva de durabilidad. Sin
embargo, mientras esta opcin fue utilizada como un encofrado permanente slo, esto no era
considerado ser un problema. Determinadas medidas de proteccin a la corrosin pueden emprenderse
para aumentar la vida de diseo de la superficie como sea necesario de acero. El hecho de que este fue un
proyecto con financiamiento privado permiti el uso innovador de acero cubierta, como un encofrado
permanente. Para un proyecto financiado con fondos pblicos, la orientacin dada en los documentos del
contrato DZHENESI sobre el uso de acero cubiertas debe seguirse para cada proyecto.
34
Table 3.1
Cost comparison between the concrete and a composite steel bridge options
Concrete bridge
Item
Steel bridge
DHC deck
Concrete decking
Steel decking
NZ$
NZ$
NZ$
$38,700
$30,200
$30,200
$9300
$9300
$9300
$113,179
$113,179
$113,179
Abutment and pier crosshead beams
$81,238
$0
$0
Abutment wing walls
$15,097
$0
$0
Settlement slabs
$12,160
$0
$0
$351,097
$390,890
$353,398
$27,620
$27,620
$27,620
$1925
$1925
$1925
$16,640
$16,640
$16,640
$0
$8400
$8400
$10,000
$10,000
$10,000
$0
$30,000
$30,000
$676,956
$638,154
$600,662
$38,802
$76,294
Preliminary and general

Earthworks
800mm diameter piles
Superstructure
Guard rails
Storm water drainage
Asphalt paving
Timber pole retaining walls
Contingencies
Concrete segmental retaining wall
Total cost
Savings
Weiti arroyo puente fue construido y terminado a mediados de 2007. Desde entonces ha aumentado el
costo del acero y posteriormente disminuy, que afectar los costos dados en la tabla 3.1. Esto sin
embargo no afecta a las lecciones aprendidas en este caso, de diseo (o al menos considerar) la
subestructura desde el principio en el proyecto y utilizando el peso baja superestructura de un puente de
acero y hormign compuesto y la alta resistencia al cociente de peso de acero.
3.4.2
SH4 Okura Realignment Project
El proyecto de reordenacin Okura demuestra que el concepto de diseo descrito en la seccin 3.5, si se
usa correctamente, ser proporcionar ahorros de costos sin importar el costo de material.
Situado en la carretera estatal 4, hacia el interior desde Whanganui, la alineacin de carretera existente
estaba sujeta a hundimientos y erosin, juntada con una alineacin de carretera sub-estndar. Para
solucionar este problema, era necesaria la realineacin de la carretera con una curva 90m North Bridge
(teniendo una constante cada transversal del 7,6%) y una recta m 96,6 largo puente sur (teniendo un 3%
Cruz cada o ancho de la lnea central del puente). El diseo conforme consisti en dos tres-puentes de
llevar la ruta reajustada, utilizando elementos prefabricadas pretensadas hormign Super T-vigas. Esta
eleccin plantea una serie de retos logsticos para el contratista incluyendo:
transportar la longitud excesiva y vigas de hormign con sobrepeso al sitio
cubierta pobre acceso a cada sitio del puente con escarpadas riberas y un ltimo puente elevado nivel
15m 20m por encima del ro (este temas creados para el posicionamiento de gras para habilitar las
vigas levantar en su lugar de nivel)
fundaciones de pila de gran dimetro hasta 2,4 m de dimetro dentro del canal del ro (acceso a la
mitad del canal del ro habra sido un problema y los tamaos de pila estaban fuera del rango de
capacidad de planta del contratista).
35
Utilizando los beneficios de un compuesto de acero y hormign mediante el sistema de cubierta de

escalera, esta forma de construccin abord los temas de transporte y construccin encontrados con el
diseo conforme; Esto incluy eliminando la necesidad de una Fundacin apilada en el ro. El reducido
peso de la superestructura del puente escala de acero cubierta permite un ahorro considerable para el
tamao de las bases necesarias para cada puente. Esto tambin incluye la eliminacin de vigas cabeza
muelle de hormign, que dio lugar a mayor ahorro en material y tiempo de construccin. Vase la figura
3.3 para detalles.
Figure 3.3
SH4 Okura Southern Bridge (image courtesy of Holmes Consulting Group)
Table 3.2 shows the cost for the South Bridge which was a straight 96m long bridge of 30m36m30m
spans. The total bridge decking area was 936m2.
Tabla 3.2 muestra resultados similares a los para el puente del arroyo Weiti descritos en la tabla 3.1; el
costo de la superestructura de ambas opciones son comparables, con la mayora de los ahorros de costos
encontramos la subestructura. Debido al peso ligero de superestructura, utilizando una opcin compuesta
de acero y hormign en lugar de una opcin concreta, el ahorro de costes en los cimientos y columnas
ascendi a un ahorro de NZ$ 334/m2 de rea cubierta. Los ahorros de costo total ascendieron a NZ$
417.456 entre las dos opciones.
Table 3.2
Cost comparison between the concrete and a composite steel bridge options
Southern Bridge (with 936m2 deck area)
Concrete bridge
Composite bridge
(NZ$/m )
(NZ$/m2)
$1054
$720
$183
$121
$1168
$1118
$2,251,080
$1,833,624
Foundations and columns

Abutments/wing walls/pier cap
Beam and deck
Total cost ($)
Savings
$417,456
36
3.5
Design concept
Para resumir los resultados del estudio descritos en la seccin 3.3, lo que fue aprendido en la seccin
3.4.1 y utilizado en la seccin 3.4.2, los siguientes factores deben ser considerados por el diseador
durante el diseo de un puente:
Acera de superestructura: utilice el menos viga principal general peso, si es posible,

independientemente del nmero de vigas, con un nmero mnimo de apoyos. Esto resulta para ser la
mejor opcin para reducir los costos de manejo y ereccin sin importar el costo del revestimiento.
Puentes escalera cubierta tambin proporcionan an ms ahorros de costos en la subestructura.
Construccin de la losa: soluciones parciales profundidad prefabricado ofrecen una solucin

rentable cuando todos los factores y limitaciones se toman en cuenta. Sin embargo, en situ de
hormign en cubiertas de acero como encofrado permanente, con la corrosin adecuada sistema
de proteccin puede proporcionar ahorros de coste adicional.
Muelles: El nmero de muelles y tamao del muelle principales vigas cruzadas pueden compararse a
un porcentaje significativo del coste total de la estructura. Utilizar el peso alto compuesto mayor
relacin de resistencia para maximizar la puente til para reducir el nmero y tamao de los muelles.
Pilares: Vuelva a colocar los pilares con muros de contencin de concretos segmentarios (o similar) y
pilares concretos o pilotes de acero.
37
Steelconcrete composite bridge design guide
Preliminary design
4.1
General
En la etapa de diseo preliminar, el diseador del puente desarrolla una solucin estructural basada en las
necesidades del cliente que se describen en la declaracin de diseo conforme a las instrucciones de
puente. Se recomienda que deben realizarse consultas con otros representantes de la industria del puente,
como fabricantes y proveedores de recubrimiento, en una etapa temprana como sern capaces de
proporcionar asesoramiento en la optimizacin del diseo del puente. Esto aumentar la eficiencia y la
rentabilidad de la opcin de puente compuesto de acero y hormign.
4.2
Design for construction
Minimizar costos pueden ser la consideracin ms obvia al embarcarse en el diseo de un puente de

carretera, la salud y la seguridad de todos los interesados en la construccin del puente y en su
mantenimiento a lo largo de su vida es responsabilidad de todos los tomadores de decisin relacin con la
adquisicin del puente. As como con el objetivo de una solucin estructural eficiente, los riesgos
asociados con el proceso de construccin deben apreciarse plenamente desde el principio.
4.2.1
Steelwork fabrication
Lneas limpias en el uso general de aspecto y mnimo de detalles complejos son ms propensos a conducir
a una estructura econmica y eficiente de puente, aunque las limitaciones externas comprometen a
menudo la seleccin de la mejor solucin estructural. La fabricacin de la yo-seccin bsica no es cara,
especialmente con el uso de moderno viga semiautomtica soldadoras (mquinas de T y yo), como se
muestra en la figura 4.1.
Total costo de fabricacin es del mismo orden que el costo del material utilizado. Con el uso generalizado
de las computadoras en el diseo y el control de mquinas de taller de fabricacin, variaciones
geomtricas, tales como plafones curvos, variando sper elevacin, plan de curvatura y precambering,
pueden lograrse fcilmente con penas razonablemente menor coste. Gran parte del costo total de
fabricacin se incurre en:
the addition of stiffeners
la fabricacin de enganches miembros
butt welding
the attachment of ancillary items (such as stud welding)
local detallando conduce a un importante manual de entrada al proceso.
38
Figure 4.1
Preliminary design
Example of a semi-automatic girder welding machine (image courtesy of D&H)
El diseador puede ejercer la libertad en la eleccin de arreglo general pero debe intentar minimizar el
nmero de pequeas piezas que deben tratarse durante el proceso de fabricacin.
La soldadura requerida, no slo para formar la viga placa sino para todos los dems componentes soldados
con autgena y conexiones en la estructura, debe ser considerada y diseada cuidadosamente. Detalles
inadecuados o la especificacin de un detalle de soldadura demasiado diseado puede resultar en un
incremento sustancial en el costo de fabricacin de la estructura.
Consejo de fabricacin debe obtenerse directamente de los fabricantes para ayudar en la eleccin de los
detalles en una etapa temprana en el diseo. La mayora de fabricantes bienvenida enfoques de
diseadores y provechoso responden a preguntas sobre mtodos de fabricacin y solicitudes de
asesoramiento sobre optimizacin.
Transporte por carretera impone ciertas limitaciones en tamao y peso de asambleas fabricadas. Ms
frecuentemente observado limitacin es una longitud mxima de 20m (incluyendo camiones y remolques);
por encima del cual notificacin especial y los procedimientos aplican. Sin embargo, los fabricantes de
puente en Nueva Zelanda se utilizan para transportar cargas largas, donde en casos excepcionales se han
transportado vigas ms este lmite. Ms orientacin sobre las limitaciones de transporte est disponible
en la gua para la gestin de vehculos pesados (DZHENESI 2006).).
39
Figure 4.2
4.2.2
Transportar un 20,3 m, 30,5 toneladas Puente viga (imagen cortesa de Eastbridge)
Erection scheme
Un esquema para la ereccin de todos los componentes principales del puente debe ser considerado
desde una etapa temprana. Acceso, mecanismos de apoyo temporal, estabilidad de la estructura erigida
parte y la necesidad de minimizar el trabajo durante cierres de carretera o ferrocarril todos pueden
tener un efecto sobre la forma y detalle de la estructura. Construccin de una superestructura del
puente compuesto generalmente procede por la ereccin secuencial de la acera, trabajando desde un
extremo al otro. Dependiendo del tipo de las tarimas usado, esto es seguido de hormigonado de la losa
de cubierta y el retiro de trabajo falsa para cubiertas in situ, o alternativamente por la colocacin de
profundidad parcial prefabricados decking seguido por un pour in situ del resto de la plataforma (vase
seccin 2.3 para ms detalles). Sin embargo, las situaciones varan considerablemente y las limitaciones
de acceso bien pueden demanda una secuencia que difiere considerablemente de la habitual. En
algunos casos las limitaciones de acceso determinar qu configuracin estructural puede ser segura y
econmicamente....
En algunas circunstancias, donde el acceso desde abajo es difcil (o imposible), puesta en marcha de uno
o ambos extremos puede ser apropiado. Si es as, esto es probable que tenga un efecto significativo
sobre la viga arreglos y detalles un arreglo de profundidad uniforme escalera cubierta es ideal para
lanzar y una proporcin menor duracin y profundidad puede ser necesarias. Consejo debe obtenerse de
un contratista experimentado.
Es mucho ms rpido establecer una conexin segura utilizando pernos de soldadura por. Uniones
soldadas in situ son ms caras, con un alto riesgo de retraso debido al mal tiempo y son onerosas para
control de calidad en un pequeo trabajo pero la soldadura puede ser considerada en los trabajos ms
grandes. Pernos o soldadura debe utilizarse durante la construccin del puente; es normalmente poco
rentable utilizar ambos mtodos.
La estabilidad de vigas durante la ereccin y bajo el peso del hormign hmedo tendr un efecto
significativo sobre el dimensionamiento de la brida superior en las regiones medio palmo y, en menor
medida, en la pestaa inferior adyacente a los soportes intermedios.
Las vigas principales de puentes de vigas mltiples a menudo se levantan en parejas apoyados; las vigas y
son ms estables que las vigas individuales y la instalacin del refuerzo de miembros a nivel del suelo es
menos peligrosa que en la altura.
Las cubiertas de la escalera estn generalmente erigida una viga en un momento (vigas principales son
generalmente de tales proporciones que puede ser levantadas individualmente, sin la necesidad de los
sistemas de restriccin temporal, como refuerzo de la cuerda a la brida superior), aunque a veces parte
til longitudes de la viga se erigen con sus vigas cruzadas ya est en marcha. Ocasionalmente, las
cubiertas completas han sido montadas cerca del sitio y transportados en posicin (generalmente
debido a las restricciones sobre los tiempos de cierre o posesin). La viga doble
4
0
4
Preliminary design
arreglo est tambin bien adaptado al lanzamiento. Durante el hormigonado, restriccin parcial de las
vigas principales contra el pandeo lateral torsional es proporcionado por las vigas cruzadas; refuerzo
adicional del plan no se suministra normalmente.
Si las vigas se erigen por lanzamiento, puede necesitarse algn refuerzo plan temporal. Tenga en cuenta
que donde las vigas principales a ser erigido individualmente requerirn torsional moderacin en los
soportes antes de la Cruz vigas estn conectadas. Debe haber espacio suficiente en los soportes
permanentes para proporcionar esta restriccin, o se necesitarn distintos trabajos temporales.
General guidance on the erection of bridge steelwork is given in the BCSA Guide to the erection of
steel bridges (BCSA 2005).
4.2.3
Slab construction
La losa de cubierta de un puente compuesto normalmente se proyectara in situ en encofrado temporal o

permanente. Tradicionalmente, Encofrado de madera, colocada entre las vigas erigidas, era ms
comnmente utilizado para toda profundidad en situ cubiertas. Recientemente, el uso de encofrado
permanente, tales como profundidad parcial prefabricados de hormign cubiertas, se ha convertido en
comn. El ltimo evita la costosa y potencialmente peligrosa operacin de despojar a encofrado temporal
despus del colado.
Profundidad parcial prefabricado decking ha sido utilizado en un nmero de escalera cubierta puentes
en Nueva Zelanda y se puede utilizar en puentes de viga mltiples tambin donde estn los rebordes
superiores de las vigas transversales en un plano comn. Tarimas de profundidad parcial prefabricado
pueden usarse para la losa se extiende hasta unos 4 metros. Se colocan en los bordes del reborde de la
viga permitiendo que los pernos de cizalla conexin a tener un slido bloque de hormign para ayudar a
transferir el esfuerzo cortante longitudinal entre el acero y el concreto. Este es el tipo de profundidad
parcial Cubiertas prefabricadas consideradas en esta gua.
Otra forma de encofrado permanente que pueda ser usado es acero trapezoidal cubierta (vanse las
secciones 2.3 y 3.4.1). Deck puede abarcar hasta 4m, permitiendo para el espaciado de la viga de la
misma longitud. Tenga en cuenta que el espesor de losa total tendr que ser superior a 250mm (mnimo
de 300mm de espesor) para ser capaces de resistir el esfuerzo cortante de penetracin causadas por
cargas puntuales de vehculo.
Es comn que no se vierta el hormign sobre toda la longitud del puente al mismo tiempo pero para
colocar concreto sobre longitudes de parte, en una serie de etapas. Esta eleccin es en parte por razones
prcticas y en parte, por hormigonado en primer lugar, las regiones medio palmo para minimizar
acaparando momentos debido a la carga. Con pilares de puente integral, los muros pantalla final
generalmente se vierten, as que no hay momentos de contencin se transfieren en el estribo debido al
peso del hormign.
Con vigas mltiples cubiertas, la losa de cubierta puede ser concretada ya sea a travs de todo el ancho
de las vigas del exteriores en cada etapa o en etapas de parte de ancho. El trabajo falso voladizo sobre la
viga exterior aplica torsin considerable a la viga exterior, dando lugar a deformaciones torsionales difcil
de predecir; los voladizos, por tanto, son a menudo emitidos tras el resto de la cubierta, especialmente si
son largos.
En la construccin de escalera cubierta la restriccin contra el giro de la viga principal proporcionada por
las conexiones de la viga transversal asegura hay rgidas restricciones al trabajo falso voladizo durante el
hormigonado y es comn a todo el ancho de la losa.
Finalmente, toda profundidad prefabricado hormign decking unidades estn disponibles; Sin embargo, hay
preocupacin por el funcionamiento de las articulaciones in situ en los Estados lmite de servicio (SLS). Estndar de
estructuras concretas NZS 3101 (SNZ 2006), tambin prohbe el uso de la conexin de un panel a comn en
localizaciones de 'alta fatiga' (como cargas de trfico en vivo, ver seccin 6.4), a menos que haya verificacin del
rendimiento. Esta conexin utiliza barras de refuerzo en forma de U de cada unidad que va desde el borde y
superpuestas
4
con las barras en forma de U de la unidad adyacente (barras rectas se pasan a travs de las interseccin
barras U para completar el refuerzo). Este detalle se muestra en la figura 4.3.
Figure 4.3
Profundidad mxima prefabricados plataforma conjunta detalle (no deseable)

U shaped rebar
(require a minimum
of D = 20db)
L dh
Straight bar
Full depth precast decking
4.3
Design for in-service maintenance
Propiedad pblica todos los puentes estn diseados para diseo de 100 aos de vida como se indica en
el manual del puente y se asumen para ser provisto de mantenimiento peridico, como se describe en las
estructuras de acero 3404.1 NZS estndar NZS 3404.1 (SNZ 2009), NZS 3101 (SNZ 2006) y la capa
estndar AS / NZS 2312 (SNZ 2002). El diseo y, en particular, los detalles, debe reconocer la necesidad
de durabilidad y facilitar cualquier mantenimiento ser necesario.
Particular issues to be addressed include:
access for repainting
detallando los requerimientos de los arreglos de drenaje para que se requiere un mantenimiento mnimo
y su fracaso no causara problemas de durabilidad
facilities for bearing replacement.
Las autoridades de cliente normalmente establecern un programa de inspecciones y mantenimiento

regulares. Acceso a reas crticas tampoco debera proporcionarse o sea posible con el mnimo de
trabajos temporales, aunque seguridad tambin debe ser considerado, para evitar el acceso no
autorizado.
Las regulaciones Resource Management Act 1991 (RMA) requieren la evaluacin de los peligros durante
trabajos de mantenimiento. El diseo debe ser como evitar o reducir, en cuanto a los riesgos siempre que
sea posible, durante el mantenimiento, minimizando el impacto ambiental o teniendo en cuenta las
preocupaciones de salud y seguridad del personal de mantenimiento.
4.3.1
Corrosion protection systems
Tradicionalmente, puentes de acero han sido protegidos contra los efectos de la corrosin mediante la
aplicacin de recubrimientos protectores. Sistemas de recubrimiento estn disponibles que va a durar
hasta 40 aos antes de su primer mantenimiento, usando productos que cumplen con los requerimientos
actuales de salud y seguridad y regulaciones ambientales.
Para asegurar la aplicacin completa y confiable y para maximizar la vida de recubrimientos protectores, el
arreglo y Detallamiento de la acera deben ser como evitar cualquier caracterstica que podra limitar el
acceso para la correcta aplicacin y mantenimiento o que hara trampa de agua y suciedad en el servicio.
En NZS 3404.1 (SNZ 2009), como se da orientacin sobre los detalles y la seleccin de un sistema de
revestimiento / NZS 2312 (SNZ
42
4
Preliminary design
2002) and El Sarraf and Clifton (2011). Figure 4.4 shows the application of a coating system onto State
Highway 25, Kopu Bridge.
Figure 4.4
4.3.2
Aplicacin del revestimiento en la viga de puente Kopu (imagen cortesa de D & H)
Weathering steel
Acero corten es una aleacin especial de acero que forma un xido estable y firmemente adherido al
carbono capa (o 'ptina') cuando sujetos a humedecimiento y secado alterno. A diferencia de la roya
comn, la ptina no se caiga de la superficie y previene la oxidacin adicional. Acero corten requiere un
mantenimiento mnimo, si los hubiere, siempre que ha sido utilizado en las circunstancias adecuadas.
Figura 4.5 muestra State Highway 1, Mercer a Longswamp rampa de salida, primer puente de acero de
meteorizacin de Nueva Zelanda, inaugurado en 2006. Se oscurecer la aparicin del acero, con una
apariencia uniforme, despus de algunos aos.
Figure 4.5
State Highway 1 Mercer to Longswamp off-ramp, shortly after construction
El uso de meteorizacin acero resulta en un costo unitario material ligeramente superior y los costes
adicionales de una 'asignacin de corrosin' a la acera (una pequea adicin al grueso es necesaria para
fines de diseo), pero no ahorra el costo de la aplicacin de una capa protectora. Los ahorros
normalmente superan a los costos adicionales como se muestra en el ejemplo de valor presente neto en
seccin 10.2.7. Consulte la gua sobre el uso del acero corten
43
to El Sarraf and Clifton (2005). Tenga en cuenta que resistiendo a la placa de acero superior a 20mm slo
est disponible desde el extranjero; Consulte el Apndice B para tamaos de placa disponible.
4.3.3
Bearing replacement
El diseo debe permitir a los rodamientos, tanto mecnicos como elastomricos, para ser reemplazadas
durante la vida de un puente, de conformidad con la seccin 4.7 del manual de puente y clusula 7.4 de la
parte 4 del estndar como 5100 puente (SA 2004). Es relativamente sencillo rigidizar la web de las vigas
principales para permitir elevacin para sustituir los rodamientos, pero debe haber suficiente espacio
(encima de las columnas, etc.) en el que se sientan los jacks. La necesidad de ayudas temporales junto a un
soporte intermedio, que a la toma de la estructura, debe evitarse, debido a los considerables costos y
riesgos que se introducen.
Secuestro bajo un diafragma de muelle (en lugar de debajo de la viga principal) debe generalmente
evitarse, a menos que crucetas integrales han sido elegidos por otras razones, porque se requiere un
diafragma ms fuerte y el detalle de la conexin, en significativo costo extra. En la construccin de la
cubierta de escalera, diafragmas de muelle pueden ser diseados para elevacin de cargas, incluso si no
son crucetas integrales, pero es difcil proporcionar elevacin arreglos para un diafragma de rodillaapoyado.
4.4
Choice of structural configuration
Para un proyecto de carretera tpica, ser la eleccin del tipo de puente entre una viga mltiples y una
escalera cubierta configuracin.
The advantages of a ladder deck configuration are:
un menor tonelaje de acero, en relacin con una cubierta de viga mltiples como resultado de
la mejor carga posible compartir en el caso de una cubierta de escalera (aunque los costos de
fabricacin por tonelada pueden ser ms altos).
adecuadas para la construccin de losa eficiente (uniformidad de espesor, fcilmente detallado para
encofrado prefabricado y ancho total colocacin de hormign)
vigas cruzadas proporcionan regular contencin rgido para voladizo de la construccin, facilitando el
uso de unidades de voladizo prefabricados
donde horizontalmente curvadas vigas son necesarios, la habitual separacin de vigas cruzadas
fcilmente proporciona la restriccin a los rebordes de fondo
Slo dos columnas necesarias en cada soporte intermedio, evitando muelles de hoja y lograr un
aspecto ms abierto
mantenimiento reducido pasivos (menos superficie de acera, menos pequeos
elementos de refuerzo y menos rodamientos)
tiempos ms rpidos de construccin proporcionan ahorros de costos
adicionales para el costo total del proyecto
. The advantages of a multi-girder configuration are:
piezas ms pequeas (de vigas principales), reduciendo as los requisitos de la gra
arriostrada pares de vigas no requieren ningn refuerzo temporal adicional para rebordes superiores
menos conexiones apernadas en sitios
distribucin de la carga bien (con el doblez transversal de la losa)
fcilmente adaptable a cualquier ancho de puente
profundidad de construccin superficial puede lograrse sin necesidad de recurrir a las placas brida
excesivamente grande.
44
Preliminary design
4.4.1.1
Influence of substructure constraints
Como se mencion en la seccin 3.5, se recomienda que el diseo de la subestructura se considera en

una etapa temprana. La forma de la subestructura en soportes intermedios, ya sea por motivos de
apariencia o la construccin, a menudo tiene una fuerte influencia sobre la forma de la superestructura.
Por ejemplo, un puente bajo autorizacin sobre tierra pobre puede utilizar vigas principales mltiples en
un solo muelle amplio, mientras que un puente de alto nivel de la misma anchura de cubierta y palmo
sobre buena tierra podra utilizar una baraja de escalera con vigas principales dobles en columnas
individuales.
4.4.1.2
Influence of skew
Puentes muy sesgadas a veces son inevitables, pero cabe sealar que el sesgo alto conduce a la
necesidad de un mayor esfuerzo de diseo, la fabricacin ms difcil y procedimientos ms complejos de
la ereccin. En particular, el modelo analtico, el detalle de vigas trimmer Pilar, pre combar y desviacin
relativa entre las vigas principales deben todos ser considerados cuidadosamente.
4.4.1.3
Influence of requirements for drainage
Drenaje de la carretera en el puente a menudo puede lograrse nicamente mediante canales de drenaje en
la cubierta del puente, pero tambin que se requiera drenaje corre por debajo de la losa de cubierta.
Arreglos para tal drenaje funciona bien puede influyen en el posicionamiento de las vigas principales en la
seccin representativa y, posiblemente, el detalle de vigas cruzadas o intermedia de refuerzo.
4.5
Preliminary sizing material selection
4.5.1
Steel
Los miembros principales de acero estructural en los puentes son generalmente grado G300 o G350 segn
AS 5100 (SNZ 2004) y las normas pertinentes de material. Grados de acero de resistencia superiores, tales
como G450L15, estn disponibles y se pueden utilizar en los diseos pero son ms caros y no todava se ha
utilizado en Nueva Zelanda en forma significativa en la construccin de puente.
Diseo de vigas de puente, el diseador debe tener en cuenta las dimensiones de la placa de acero
disponibles para minimizar el desperdicio y optimizar el tamao de la viga. Por ejemplo, mediante la
reduccin de la cantidad mxima de bridas de una placa para minimizar el desperdicio esto ayudar a
mantener bajo el costo de fabricacin. Otro consejo es elegir la profundidad de web basada en el ancho
de la placa disponible. Por lo tanto, si el diseador requiere un haz de 1200mm profundo que abarca
18m, con las bridas 750mm de ancho y 32mm de espesor, y la web 16 mm, una placa gruesa de 1530mm
ancho 32mm es elegida para las bridas. Esta placa se corta a la mitad de las pestaas y una placa de
1532mm ancho 16mm es elegida para la web. Para lograr el lapso requerido 18m, una placa de 6m y 12m
para esas dimensiones es elegida y butt soldadas con autgena juntas. Tenga en cuenta que cuando la
placa se corta, el diseador debe compensar la prdida de 10mm para cada corte y tomar en cuenta el
subsidio precamber. Se recomienda que los diseadores discutan este asunto con los fabricantes para
asegurar....
Las dimensiones disponibles de Nueva Zelanda hizo placas figuran en el Apndice B.1 y placas importados
figuran en el apndice a.2. Tenga en cuenta que la longitud mxima de la placa es normalmente de 12m y
hay un lmite en el peso de una placa individual (es decir, en la combinacin de largo, ancho y espesor)
que puede limitar an ms la longitud de placa muy gruesa.
Para refuerzo de miembros, seccin laminada en caliente ngulos, canales y secciones pueden utilizarse, especialmente los
ngulos para K - o X-ortodoncia. Si se va a utilizar acero corten, no hay disponibles secciones laminados en caliente; Sin embargo,
los fabricantes pueden fabricar secciones similares de placa; Busque el Consejo de un fabricante sobre qu secciones son
econmicamente factibles. Acero corten relacionados con la orientacin se encuentra en El Sarraf y Clifton (2005).
45
Seleccin de la subrasante particular (para la dureza) generalmente no necesita ser considerados en el

diseo inicial a menos que la disponibilidad es probable que sea un problema, como grueso no estndar o
dimensiones (vase el Apndice B para obtener ms informacin de dimensiones y espesores de placa
estndar). Orientacin sobre la disponibilidad y tiempos de entrega para productos especficos y tamaos
debe obtenerse tan pronto como sea posible para optimizar el diseo de viga de puente final de
fabricantes o proveedores de acero. Si el perodo de construccin se limita, tiempos de entrega pueden
afectar la eleccin de los componentes.
4.5.2
Concrete
La eleccin del grado concreto depende de su funcin. Por encima de componentes de la subestructura
tierra como pilar vigas, vigas principales de muelle y muelles, 40MPa concreto es comn. Para cubiertas de
losa, 50MPa generalmente se especifica para cubiertas prefabricadas y cuando regulan aspectos de
durabilidad (es decir el puente se encuentra en un ambiente marino), mientras que 40MPa concreto se
especifica para Deck in situ y rellenado.
Todos los componentes concretos, ya sea in situ o prefabricado, son fabricados de acuerdo con NZS
3104 especificacin para la produccin de hormign (SNZ 2003), construido para construccin NZS 3109
concreta (SNZ 1997) y diseados para NZS 3101 (SNZ 2006).
4.6
Preliminary sizing multi-girder bridges
4.6.1
Girder spacing
Los momentos transversales totales de la losa de una cubierta de viga mltiples no son particularmente
sensibles al espaciado de la viga en el rango de 2,5 m a 3,8 m (el incremento de momento local con el
palmo es ms o menos equilibrado por una reduccin en el momento en que surja de la transferencia de
carga de una viga a la siguiente). Es ventajoso para elegir la viga ms amplia espacio como sea posible,
compatible con otras consideraciones geomtricas y con la opcin apropiada del decking, tales como
profundidad parcial prefabricados decking descritos en la seccin 2.3. Es preferible utilizar un nmero
par de vigas para que ellos puedan combinarse durante la construccin, pero puede utilizarse un nmero
impar, si se adeudan se prevn para el montaje de las vigas individuales.
Al seleccionar un espaciamiento adecuado de la viga, debe prestarse atencin a los voladizos en los
bordes de la cubierta. La longitud del voladizo desde el eje de viga exterior normalmente debe
restringirse a cerca de 1,5 m a 2m, incluyendo la viga de borde.
El diseo para la carga de trfico accidental aplicada sobre la barrera, la longitud de voladizos puede
necesitar ser restringida, y la losa aumento en grosor localmente.
4.6.2
Girder profile
Por simple se extiende sobre unos 25 metros, una profundidad de construccin (parte superior de la losa
en la parte inferior de la viga) de entre aproximadamente 1/18 y 1/30 del tramo puede lograrse con vigas
prefabricadas; Aunque la solucin ms econmica ser hacia el extremo ms profundo de esta gama.
Tenga en cuenta que el costo de la aplicacin de un sistema de capa protectora debe ser considerado
como el costo puede aumentar con la mayor superficie de una viga ms profunda en comparacin con un
rayo ms corto. Por ms cortas, la profundidad es probable ser proporcionalmente mayor,
particularmente para los palmos debajo de 20m. Esto puede limitar el uso de una viga soldada con
autgena debido a la profundidad disponible en comparacin con una viga prefabricada.
Para compuesto continuo se extiende con bridas paralelas, la profundidad de la construccin (otra vez,
desde la parte superior de la losa en la parte inferior de la viga) es tpicamente entre 1/20 y 1/25 del
mayor rango. El uso de ancas curvas o cnicos puede reducir la profundidad de la construccin en medio
palmo, a expensas de mayor profundidad en los apoyos internos. Una seleccin de arreglos tpicos se da
en la figura 4.6.
4
Preliminary design
4.6.3
Flange and web sizes
Experiencia y unas reglas del pulgar puede utilizarse a menudo una seleccin inicial de tamaos. Como
consecuencia del peso de la acera contribuyendo poco a momentos de diseo global, la seleccin puede
ser rpidamente refinada. Un tal conjunto de normas simples se da en el Apndice C.
Vanos que deben fabricarse con ms de una sola longitud a cada viga principal dan la oportunidad de
variar la composicin de la viga en las diferentes piezas requeridas para cada tramo. Longitud mxima de
las piezas est influenciada por el transporte (cargas de ms de 20m de largo requieran arreglos
especiales) y por la longitud de la placa que est disponible.
Figure 4.6
Proporciones palmo/profundidad tpica para continuo se extiende en un puente de vigas mltiples
L/20 to L/25
L/30 to L/40
L/15
L/18
L/40
Note: Depth measured to top of slab
4.6.4
Slab thickness
Para el diseo inicial, elija un espesor de losa de 250mm. Esto puede ser refinado en el diseo detallado.
Si la barrera del voladizo est diseada para una carga de trfico accidental, usar un espesor inicial de
330mm (para un voladizo de 2m) en la raz del cantilever, reduciendo a 250mm en la primera viga
principal interna.
47
4.7
Preliminary sizing ladder decks
4.7.1
Girder spacing
El espaciado de la viga principal debe ser tal que voladizos son unos 1,2 m a 2m de largo. Las vigas
principales son tpicamente espaciadas entre 5,5 m y 18m de distancia para adaptarse a la anchura del
camino. Para una ms amplia cubierta, utilizar dos bandejas de escalera separada o dos conjuntos de
acera escalera cubierta con una losa comn. Si barreras de contencin muy alta tienen que ser
proporcionado en el borde de la cubierta, elija un voladizo corto o espesar la losa.
Elija un espaciado viga transversal entre 3,3 y 4 metros. Espaciamiento hasta 4m se logra usando
profundidad parcial prefabricado terrazas como encofrado permanente. Tenga en cuenta que si el puente
es curva en plan, el espaciado de la viga transversal ser mayor en el exterior de la curva y longitudes de
encofrado variar el ancho de la cubierta.
4.7.2
Girder profile
Vigas principales son generalmente de profundidad uniforme, aunque haunched vigas pueden adaptarse
a algunas situaciones. Para vigas de profundidad uniforme, la profundidad total (viga + losa) debe ser
normalmente entre 1/15 y 1/25 del rango de mayor (1/25 puede aparecer muy delgados). Para las
cubiertas de ancha, la profundidad debe ser hacia el extremo ms profundo de la gama.
Vigas transversales deben tener una profundidad de entre 1/12 y 1/18 del tramo entre las vigas
principales. Generalmente tienen un reborde inferior recta, pero la brida superior normalmente seguir el
perfil transversal de la calzada.
4.7.3
Flange and web sizes
Elegir tamaos brida inicial sobre la base de la experiencia previa o el uso de modelos de viga de lnea
simple (la proporcin de la carga llevada por cada viga se determina fcilmente por una distribucin
'esttica' transversalmente).
Para tramos largos, las bridas pueden ser bastante gruesas, hasta a 80mm espesor o incluso 100mm de
espesor. El uso de un mayor grado de resistencia (superior a G350) puede ser apropiado en algunas
circunstancias, pero son mayores grados de resistencia, en la actualidad, significativamente ms caro y
menos fcilmente disponibles; tiempo de entrega tambin es probable ser ms largo. Grado G350 se ha
utilizado casi exclusivamente para los puentes de escalera cubierta en Nueva Zelanda.
Para vigas, cruzadas secciones laminada en caliente o Vigas soldadas pueden utilizarse. Elija una viga
placa vigas de seccin sobre la base de las vigas transversales actuando como simplemente apoyado.
4.7.4
Slab thickness
Elija un espesor de losa de 250mm para una viga transversal espaciando hasta 4.0 m aproximadamente,
con el debido respeto a cubierta necesaria para una mayor durabilidad (y al grado del concreto, que es
normalmente 40 MPa). Donde llevar a una barrera de contencin de muy alto nivel, habr grandes
momentos y fuerzas laterales (fuerzas externas, provocando tensin en la losa) para sostenerse, con
consecuencias en el espesor de losa. Para lograr un voladizo de 2 m portando una barrera semejante,
puede necesitarse un espesor de losa de hasta cerca de 330 milmetros. Este espesor necesitar ser
afilados a la regular losa interior del espesor de la viga exterior sobre una longitud de 1m a 2m. Tambin
debe prestarse atencin a los medios de transferencia del momento transversal en los extremos de las
vigas transversales.
48
Design standards
Design standards
The responsibility for the maintenance of state highway roads and bridges in New Zealand rests with the
NZTA, while territorial authorities (such as city and district councils) are responsible for the maintenance
of all non-state highway roads and bridges.
The Bridge manual defines design loadings, load combinations and load factors, together with criteria for
earthquake resistant design. It does not, however, define detailed design criteria for the various materials,
but refers to standards such as those produced by Standards New Zealand, Standards Australia and the
British Standards Institution. The standards to be used should be the editions referenced in the Bridge
manual, including all current amendments. For cases when specific portions of these standards are
referred to in the Bridge manual, any reference in these standards to specific loads or load combinations
that might conflict with the provisions in the Bridge manual should be disregarded.
For the purpose of assessing probabilistic effects of loading such as wind, earthquake, flood and live load
fatigue, and for consideration of long-term effects such as corrosion, creep and shrinkage, the design
working life of a bridge or an earth retaining structure is taken in the Bridge manual to be 100 years. This
may be varied by the controlling authority if circumstances require it. It should be noted that the 100-year
design working life exceeds the minimum requirement of the New Zealand Building Code.
This chapter reviews briefly the range of documents that are used in the design and construction of
bridges in New Zealand. These documents form the basis of the design guidance given in this publication.
5.1
The Bridge manual
The Bridge manual is the main bridge design document that outlines the criteria for the design of new
bridges and evaluation of existing bridges in New Zealand. It comprises the following principal sections:
Section 1: Technical approval and certification procedures
Section 2: Design general requirements
Section 3: Design loading
Section 4: Analysis and design criteria
Section 5: Earthquake resistant design of structures
Section 6: Site stability, foundations, earthworks and retaining walls
Section 7: Evaluation of bridges and culverts
Section 8: Structural strengthening
Appendices A to G
All vehicle traffic loading and design loading requirements are based on section 3 of the Bridge manual.
Earthquake loadings are given in section 5 of the Bridge manual, which incorporates parts of the guidance
given in part 5 of NZS 1170.5 (SNZ 2004e). Future amendments and revisions to the Bridge manual are
planned and the designer should comply with these updates.

For steel and composite construction, section 4.3 of the Bridge manual currently refers to NZS 3404 Steel
structures standard (SNZ 1997b) and AS 5100.6 (SA 2004). Box girders are outside the scope of this
document.
49
For reinforced concrete and prestressed concrete, section 4.2 of the Bridge manual refers to NZS 3101
Concrete structures standard (SNZ 2006).
5.2
AS 5100
The Australian AS 5100 Bridge design standard (SA 2004) is composed of seven parts and provides a
comprehensive design standard specific for bridges, unlike the New Zealand NZS 3404 Steel structures
standard (SNZ 1997b), which provides guidance for all steel structures. AS 5100, part 6 (SA 2004a), is the
main design standard that will be used in this guide for the design of steel-concrete composite bridges.
The seven parts are:
Part 1: Scope and general principles
Part 2: Design loads
Part 3: Foundation and soil-supporting structures
Part 4: Bearings and deck joints
Part 5: Concrete
Part 6: Steel and composite construction
Part 7: Rating of existing bridges.
It should be noted that AS 5100 (2004) references BS 5400 (all parts), in addition to a number of
Australian and international documents. For the reasons stated below and for the sake of consistency,
most of the design guidance on composite construction given in this guide will be based on the relevant
parts of AS 5100 (SA 2004) and its 2010 amendment no. 1.
5.3
NZS 3404
There are currently two versions of the New Zealand Steel structures standard in print: NZS 3404.1 and 2
(SNZ 1997) which includes two amendments dated June 2001 and October 2007, and NZS 3404.1 (SNZ
2009). The latter specifically provides updated guidance on materials, fabrication and construction, which
includes corrosion protection and inspection of welding and bolting, while the former also includes
general design guidance of steel structures.
Section 4.3 of the Bridge manual specifies that NZS 3404 (SNZ 1997b) should be used for the design of
steel componentry of bridge substructures together with any seismic load resisting componentry that is
expected to behave inelastically. Steel componentry of bridge superstructures, including seismic load
resisting components expected to behave elastically, should be designed in accordance with AS 5100.6
(2004).
NZS 3404.1 (SNZ 2009) is the first part of seven, which in time will supersede the 1997 version. Designers
must be aware of the two versions and the superseding parts that will be developed and published in the
coming years. The 2009 version is mainly referenced in this guide for providing guidance on the
fabrication of steel in bridges (section 8.1 of the standard).
50
Design standards
5.4
NZS 3101
NZS 3101 (SNZ 2006) provides guidance on the design and use of concrete, which includes bridge-related
guidance especially for concrete decking, concrete bridge fatigue design, as well as precast concrete
components in the bridge superstructure. Important durability design guidance is also given.
5.5
BS 5400
The British standard BS 5400 Steel, concrete and composite bridges consists of 10 parts, which were
published and revised from 1978 through to 2010. In March 2010, BS 5400 was formally superseded by
the Eurocodes and withdrawn. Although the Bridge manual no longer references BS 5400, this standard
still remains useful as a reference document.
5.6
Eurocodes
The Eurocodes are a set of structural design standards that have been developed by CEN, the European
standards body, and adopted throughout Europe. The Eurocodes are published in a total of 58 separate
parts, each dealing with either general, material-specific or structure type-specific matters. Each part
contains principles and application rules that are common in all the adopting countries, apart from certain
aspects that have been left for national choice, as determined by the individual national standards body.
The Eurocodes are published unchanged by each national standards body, together with a national annex
that implements the Eurocode in the country and gives the national choices for that country.
The Eurocode parts that are relevant to the determination of resistance and serviceability of steel-concrete
composite bridges are:
Eurocode 3 part 1-9 EN 1993-1-9 (CEN 2005a)
Eurocode 3 part 2 EN 1993-2 (CEN 2006)
Eurocode 4 part 2 EN 1994-2 (CEN 2005b)
The use of rules from the appropriate Eurocode will be considered in this document where guidance has
proved to be insufficient in AS 5100 and the other stated standards. This is especially important to the
fatigue design of the composite bridge structure. Reference is made, for convenience, to the UK
publications, for example to BS EN 1993-2, and the National Annex to BS EN 1993-2. This is especially
useful as it is easier to obtain a copy of the UK version of the Eurocode from the British Standards
Institution (BSI), in addition to the fact that they are printed in English.
In addition, BSI has issued several published documents such as PD 6695-2 (BSI 2008) that provide noncontradictory complementary information for use with the Eurocodes. Where relevant, such documents are
referenced in this guide.
5.7
Basis of design
The design philosophy to be adopted for bridges is set out in clause 2.1.1 of the Bridge manual: Highway
structures shall be designed to satisfy the requirements of both the ultimate and serviceability limit states
when acted upon by any of the combinations of loading defined in this document. Clause 2.1.2 defines
the requirements at the two limit states:
51
5.7.1.1
Ultimate limit state (ULS)
The state beyond which the strength or ductility capacity of the structure is exceeded, or when it cannot
maintain equilibrium and becomes unstable.
In AS 5100.6, clause 3.2 expresses the strength requirement as:
S* Ru
In which (R ) is the design capacity of the structure and its component members and (S*); is the design
u
action effect (ie the internal forces and moments due to the design loading).
5.7.1.2
Serviceability limit state (SLS)
The state beyond which a structure becomes unfit for its intended use through deformation, vibratory
response, degradation or other operational inadequacy.
In AS 5100.6 (SA 2004) clause 3.3 sets out more detailed SLS requirements. For actual limits for deflection
etc reference must be made to NZS 3101 (NZS 2006).
5.8
Project-specific requirements
For state highway bridges, the NZTA is responsible for defining the design parameters and endorsing the
final design for construction under the NZTA technical approval process. Details of this process are given
in section 1 of the Bridge manual.
Often owners of private bridges, not managed by either the NZTA or the territorial authority, choose to
comply with the provisions given in the Bridge manual, since this is the main bridge design document in
New Zealand.
5.9
Health and safety regulations
Adems de los requisitos tcnicos de los documentos mencionados, las consideraciones de salud y
seguridad, tales como los establecidos en la salud y seguridad en la ley de empleo de 1992 (ley de HSE) y
en una casa de salud y seguridad poltica, conducen a los requisitos para una evaluacin de los riesgos en
todas las etapas de la construccin. El objetivo debe ser eliminar o reducir a un nivel aceptable, los riesgos
identificables. Los diseadores son por lo tanto obligados a anticipar cmo se construir una estructura y a
evaluar los riesgos involucrados en la construccin. Deben hacer arreglos para que toda la informacin
pertinente que se deben transmitir a los dems que debe trabajar sobre la estructura incluyendo los
propietarios.
It is recommended that health and safety consideration and risk assessment during the conception and
design of the structure, during its use, repair and its final removal/demolition should also be considered.
Guidance on these matters can be found in the HSE Act, AS 3828 (SA 1998), or the UK CDM Regulations
(Health and Safety Executive 2007).
52
6
Calculation of action effects
6.1
Structural analysis
For steel and composite construction, section 4 of AS 5100.6 (SA 2004) states that one of the following
two methods shall be used to determine the design action effects in a structure:
elastic analysis
rigorous structural analysis.
Section 4 also gives additional requirements for buckling analysis, analysis of composite beams, staged
construction, connections and determination of longitudinal shear.
6.1.1
Elastic analysis
Clusula 4.2 de AS 5100.6 (SA 2004) describe los requisitos para el anlisis elstico, donde se supone que
los miembros individuales permanecen bajo la accin de las cargas de diseo en el ltimo y el
mantenimiento de los Estados lmite elstico. Dos tipos de anlisis estn contemplados en la clusula 4.3:
anlisis de primer orden y segundo orden.
Anlisis elstico de primer orden no tienen en cuenta los cambios en la geometra de la estructura o los
cambios a la rigidez efectiva de miembros debido a fuerzas de compresin. El cambio en la geometra es
suficiente para modificar las fuerzas internas y momentos, los cambios se refieren a los efectos de
segundo orden.
Efectos de segundo orden debido a la resistencia a la compresin axiales cepas en los miembros
individuales que conducen a desplazamientos conjuntos se refieren comnmente como P- effects;
efectos debido a la flexin de los miembros individuales se denominan P- effects. Cabe sealar que cepas
de resistencia a la traccin tambin pueden causar efectos de segundo orden y conducir a una reduccin
en los momentos de primer orden y tijeras pero esto normalmente se descuida en el diseo, con excepcin
de estructuras trianguladas como cerchas. Efectos de segundo orden son especialmente importantes para
el diseo ssmico; Esto se discute brevemente en la seccin 6.5.
Anlisis elstico de primer orden es adecuado para la mayora de las estructuras del puente compuesto.
Donde es necesario considerar los efectos de segundo orden, clusula 4.2.1.2 de AS 5100.6 afirma que
pueden ser determinados por una simple amplificacin de los efectos de primer orden o un anlisis elstico
de segundo orden. Para la primera opcin, anlisis elstico de primer orden da los valores del diseo
mximo calculado momentos de flexin (M *) en los miembros. Estos son amplificados por el factor de la
amplificacin de momento ( o
), que es m b s determinada de conformidad con las clusulas 4.2.2.2 y
4.2.2.3 (y basado en miembro pandeo anlisis, vea la seccin 6.1.3), para dar el diseo doblado momentos
(M *). Por la segunda opcin, clusula 4.2.1.2 se refiere a un anlisis elstico de segundo orden segn el
Apndice C del AS 5100.6. Un anlisis elstico de segundo orden completo determinara (M *) directamente
y evitar la referencia a la clusula 4.3.
6.1.2
Rigorous structural analysis
Clause 4.10 of AS 5100.6 gives general requirements for a rigorous structural analysis (as an alternative to
an elastic analysis) but gives very little detailed guidance. Rigorous structural analysis takes account of
geometric deformations and non-linear behaviour of materials but would not be used for conventional
bridge design.
6.1.3
Member buckling analysis
Clusula 4.3 de AS 5100.6 establece las reglas para la determinacin de las cargas de pandeo elsticas de
miembros. Esto se utiliza para determinar el factor de amplificacin de momento (
) en el apartado 4.2.2.3
del AS 5100.6. Para m compuesto puentes, miembro anlisis de pandeo se utiliza para determinar la
longitud efectiva de los miembros de compresin en los sistemas de arriostramiento, que es requerido por
la clusula 10.3.2.
6.1.4
Analysis of composite beams, girders and columns
Clusula 4.4 de AS 5100.6 da requisitos adicionales para el anlisis de vigas compuestas, vigas y
columnas. Proporciona reglas para la determinacin de las propiedades de seccin eficaz, como los
factores relacionados con el hormign y la informacin necesaria, al analizar a un miembro compuesto de
acero y hormign. Incluye la consideracin de esquileo del retraso en la losa de hormign y la diferencia
en el mdulo de elasticidad de concreto y acero. Entonces se utiliza con las disposiciones de NZS 3101
(SNZ 2006) en el diseo de la losa de hormign de la estructura del puente, vea las secciones 7.1 y 7.10
para ms detalles.
La metodologa para determinar los momentos longitudinales y cizallamiento asociado y reacciones en una
viga continua compuesta es dada en el apartado 4.4.3; Esto incluye orientacin sobre teniendo en cuenta
de hormign que est roto en tensin. Clusula 4.4.4 da los requisitos para el clculo de las desviaciones
en un miembro compuesto: para ello, puede suponer estar intactas concreto en tensin.
6.1.5
Staged construction
Clusula 4.6 de AS 5100.6 da los requisitos adicionales para el anlisis de la construccin puesta en
escena del puente compuesto de acero y hormign. Afirma que adecuacin cheques son necesarios para
cada etapa de construccin, tales como cuando la seccin de acero inicialmente lleva su propio uno
mismo-peso y el peso de la cubierta de hormign, pero acta entonces compositely con la losa de
hormign para cargas aplicadas posteriormente. El resultado es diferente de flexin rigidez y carga la
distribucin en cada etapa de la construccin.
Anlisis independientes son necesarios, que representa cada etapa de construccin sucesivas. Esta serie
de anlisis se siga la secuencia de hormigonado y tener en cuenta la distribucin del peso de mojado de
hormign, especialmente la de los voladizos. Ser una serie de estructuras parcialmente compuestas.
Variables acciones (cargas de trfico) se aplican a una estructura completamente compuesta, generalmente
con propiedades compuestas a corto plazo.
Clusula 4.6.2 distingue entre secciones compactas y no compacta en el tratamiento de ULS los efectos
debido a las diversas etapas de construccin. Para las secciones compactas, los efectos totales de cada
seccin (momentos y tijeras) simplemente se agregan juntos para la verificacin de la capacidad de
diseo. Para no-compact secciones las tensiones en cada seccin para cada etapa de construccin se
agregan juntos para dar totales tensiones, para la verificacin de la capacidad de diseo. Clusula 4.6.3
requiere que, en SLS, independientemente de la clasificacin de la seccin, las tensiones son
determinadas para cada etapa y suman (en la prctica esto slo tiene que ser aplicado para las secciones
compactas, para asegurar que limitar las tensiones no se sobrepasen para las secciones no-compact,
verificacin de ULS es siempre ms onerosas).
Por lo general, hay dos veces tantas etapas como palmos, porque concreto se coloca sucesivamente en
cada una de las regiones medio palmo, seguidas por las regiones restantes en cada soporte. Cuando los
voladizos se concret en una etapa diferente de la anchura principal de losa, esto deber tenerse en
cuenta en los anlisis.
6.1.6
Connections
Clause 4.7 of AS 5100.6 gives the additional requirements for the analysis and design of connections
whether they are bolted or welded. Each element of the connection shall be designed so that the structure
54
6
is capable of resisting all the applied design actions. The design capacities of each element shall not be
less than the calculated design actions effects. The connection design parameters are given in clause 4.7.1
and subsequent clauses.
6.1.7
Longitudinal shear
Clause 4.8 of AS 5100.6 gives the requirements for calculating the design longitudinal shear force per unit
length on a shear plane. This will be used to determine the number of shear connectors needed. Guidance
on calculating the transformed concrete area can be found in Beer et al (2008).
6.1.8
Shrinkage and differential temperature effects
Clause 4.9 of AS 5100.6 states how the effects of shrinkage modified by creep and differential
temperature are to be evaluated. Requirements for evaluating the consequent effects on the longitudinal
shear forces under both ULS and SLS are also given.
6.2
Computer modelling
Modelizacin informtica ahora comnmente se utiliza para ms anlisis global del puente, incluso para
simplemente apoyados solos palmo puentes. Esto es especialmente til para vigas varios puentes, donde la
distribucin de los efectos de la accin entre las vigas sera difcil determinar con precisin a mano. Potente
software analtico se est convirtiendo en disponible que no slo ayuda a los diseadores en la
determinacin de los efectos de la accin sobre la estructura, sino tambin en la determinacin de los
momentos crticos de pandeo de los miembros y los valores de las estructuras de pandeo elstico.
Cabe sealar que no importa cun avanzado ordenador se convierten en los programas de modelado,
siempre es recomendable que el diseador novato puente (o incluso experimentados diseadores)
verifiquen los resultados obtenidos por el programa para asegurar que se calcularon los datos correctos.
Esto puede ser tan simple como el clculo de la mano el momento total y acciones de corte actuando en
todo el puente de estructura. El propsito de este ejercicio es para acumular experiencia del diseador y
para adquirir un 'toque' de lo que deberan ser los resultados correctos. Esto proporciona un conocimiento
que permitira identificar errores en las salidas del modelo de computadora y en funcionamiento alrededor
de algunas de las limitaciones que puede imponer el programa.
Al final, depende de la diseadora para determinar el programa de modelado informtico que procede al
anlisis y diseo de la estructura de un puente. Una breve introduccin de la computadora disponible
diferente opciones de modelado se resumen abajo.
Note that, if second-order analyses are being used to determine second-order effects directly, then the
computer software must be capable of modelling both P- and P-effects. Additional guidance is
available in appendix E of NZS 3404 (1997).
6.2.1
2D grillage models
6.2.1.1
Multi-girder decks
Para las cubiertas de puente de vigas mltiples, un emparrillado 2D simple dar resultados adecuados para
puentes no integral (ver (Iles 2010) para orientacin sobre modelado integrales puentes y (Hambly 1990)
para orientacin adicional). En estos modelos, la estructura es idealizada como una serie de elementos viga
longitudinal y transversal en un solo plano, rgidamente interconectados en los nodos. Los viga transversal
elementos pueden ser ortogonales o sesgada respecto a las vigas longitudinales. Cada elemento rayo
representa una seccin compuesta (por ejemplo viga principal con losa asociada) o un ancho de la losa (por
ejemplo un elemento transversal puede representar un ancho de losa igual a la separacin de los
elementos transversales). Donde los soportes estn en ngulo recto a las vigas principales, se utiliza un
emparrillado ortogonal. Donde los soportes tienen una pequea inclinacin, el emparrillado puede ser
sesgada (las lneas
55
de refuerzo tambin probablemente ser oblicua en estos casos). Cuando la inclinacin sea superior a 20
grados, un sesgo emparrillado tiene dificultad en modelar el comportamiento de la losa; se utiliza un
emparrillado ortogonal con extremos sesgados (pero entonces un modelo local tambin se puede necesitar
en los ngulos obtusos porque el modelo de emparrillado no puede separar los efectos torsionales
llevados por la losa de los llevados por la viga de deformacin, que es una preocupacin particular en esas
regiones). Ejemplos de todas las configuraciones de tres se muestran en la figura 6.1.
Debe proporcionar una lnea de elementos a lo largo de cada viga principal; lneas intermedias
(representando a losa solamente) se utilizan para refinar la malla para que los efectos de la carga de las
ruedas pueden ser evaluados (en lugar de desarrollar un modelo separado o refirindose a las superficies
de influencia placa estndar para determinar los efectos locales). Una viga de borde se proporciona
generalmente para facilitar el modelado de los voladizos. Porque los elementos de la viga transversal no
representan elementos estructurales discretos, el espacio puede ser elegido por el diseador para facilitar
el anlisis. En general, el espacio no debe exceder de 1/8 del palmo para modelar los efectos globales.
Espaciado uniforme nodo debe elegirse en cada direccin cuando sea posible, aunque puede ser til
localizar los nodos en las posiciones del empalme o refuerzos (de modo que los valores de momentos y
cizallamiento en estas posiciones estn disponibles en la salida). Para puentes de viga mltiples, cizalla lag
es improbable reducir la anchura eficaz de la losa por debajo de su anchura real (vase la referencia....
Figure 6.1
2D grillage models for a three-span multi-girder bridge
a) Orthogonal grillage
b) Grillage for spans with small skew <20
c) Grillage for spans with large skew >20
Section properties for the composite main beams in the sagging moment region should use the full
composite second moment of area; if there are intermediate longitudinal lines, the elements should be
given only the properties of the slab itself. For continuous composite main beams in the hogging moment
region, clause 4.4.3 of AS 5100.6 specifies that the stiffening effects of the concrete over 15% of the
length of the span on each side of the internal support may be neglected. In this case, the properties of
the steel section and the longitudinal reinforcement should be used. Section properties for transverse
beam elements representing the slab alone should use a width equal to the element spacing.
In regards to the torsional stiffness of the slab, it can be conservatively taken as zero in a 2D grillage
when modelling the slabs, thereby designing them for bending effects only. Even though in reality, the
torsional stiffness of the slab will assist in distributing the loads between the transverse and longitudinal
beams. Intermediate bracing (between beam pairs) should be modelled, because it does affect the local
transverse bending stiffness (although it does not significantly affect distribution of load between main
beams). A shear-flexible member should be introduced (derived from a local plane frame model of the
56
bracing); this will give rise to local forces that can be used to verify the adequacy of the web stiffener to
flange connection. (The inclusion of bracing in the model gives a better distribution of effects than the
alternative of applying deformations from an unbraced model to a local plane frame.)
6.2.1.2
Ladder decks
For ladder deck bridges, 2D grillage models can be used, although they are not fully able to model the
local effects of the deflections of the cross girders and their interaction with the lateral bending of the
main girder bottom flanges. For cross girders, the appropriate width of slab acting with the cross girder is
the spacing of the cross girders (ie one half of the distance to the next girder on either side) as specified
in clause 4.4.1 of AS 5100.6. However, further guidance related to ladder deck cross girders is also found
in clause 5.4.1.2 of EN 1994-2 (CEN 2005) which states that the slab width should also not be more than
one quarter of the spacing between main girder webs plus the spacing to the outer studs on the cross
girder.
Joints in the grillage model should be rigid connections. This applies not only to the joints between
elements along a beam (which must be rigid) but also to the joints between cross girders and main girders
(for both the bare steel/wet concrete condition and the composite condition), assuming that either a
lapped or spliced connection is used.
The same guidance on torsional stiffness is taken as zero in ladder deck bridges and the slab is designed
for bending effects only. In reality the slab and the lateral bending stiffness of the bottom flanges of the
main girders provide additional restraint against the twisting of the main girder (in effect this is warping
restraint) but this is too complicated to model in a simple 2D grillage.
In many cases, the bridge will be straight and the cross girders square to the main girders. An orthogonal
grillage model is well suited to this arrangement. Where the bridge is skew, the cross girders will normally
still be square to the main girders although the spacing may have been adjusted locally to the supports so
that the bearings are below cross girder to main girder connections.
Figure 6.2 shows a typical 2D grillage model suitable for the global analysis of a three-span ladder deck
bridge. The slab mesh should be at about 3m spacing transversely and half the cross girder spacing.
Cantilever slabs and edge beams should also be modelled. (This figure shows coincident cross girders and
trimmer girders at the obtuse corners but this may not be practical, as mentioned in section 2.2.3; careful
attention should be given to modelling in these corners.)
Figure 6.2
2D grillage model for a skew three-span ladder deck
Elements representing main girders, cross girders and trimmer girders.

Elements representing deck slab.
6.2.2
3D grillage models
For ladder deck bridges, a 3D grillage skeleton model, essentially the same plane model as a 2D model
but with vertical elements at every cross girder connection, connected at the bottom to beam elements
57
representing the bottom flanges, is better able to represent the interaction between cross girders and
main girders. The vertical elements have negligible bending stiffness in the plane of the web but have the
stiffness of the effective stiffener section out of that plane. Vertical bending stiffness of the main girders is
assigned wholly to the upper members and the bottom flange elements represent only the plan bending
of those flanges. The differential loadings on adjacent cross girders generate different deformations in the
adjacent U-frames and thus plan bending of the bottom flange; additional guidance is available in
appendix A of AS 5100.6.
Although this type of model can be limited to portions of the deck, to determine local effects that reflect
the interaction between cross girder U-frames (and thus is used in conjunction with a coarser global 2D
grillage), once the extra complexity is addressed it is probably better to model the whole bridge in this
way. The one model then determines both global and local effects (although it is not possible to separate
global and local effects).
Finally, designers must be aware of some of the limitations that may be experienced when undergoing a
grillage model, whether it is 2D or 3D. Some of these limitations are:
Eccentricity among the structural elements of a bridge cannot be taken into account in the model.
Inevitability, additional internal forces and possible load distributions are ignored.
It is difficult to take into account torsion and distortional warping effects.
It is difficult to investigate the buckling phenomena of the steel girders during erection stages and the
deck concreting.
Diaphragms, bracing systems and stiffeners, which are usually installed in order to improve global and
local stability and increase torsional rigidity, cannot be adequately represented. These limitations may
be addressed by the use of other available software, which may be more complex.
6.2.3
Finite element models
A full linear 3D finite element model can give a more realistic determination of the structural response,
particularly for ladder decks. Finite element programs are currently the most powerful and versatile
analytical tool available. The main advantage finite element modelling provides is that it can be applied to
any structure regardless to the level of complexity. For bridges, models can be built using plate elements
and beam elements. Plate elements are used for the deck slab (using the Wood Armer Method to analyse
the slab) and for the webs of the main girders (and for the webs of cross girders in ladder decks). Beam
elements can be used for the flange plates (aligned with their bending stiffness in the plane of the flange
and, for the top flange, having an offset from the slab elements), for web stiffeners (representing the
effective Tee section) and for triangulated bracing.
Designers need to be aware of the capabilities of the different types of elements, how to assign
appropriate properties and how to interpret the results of the analyses. For example: shell elements will
automatically account for shear lag to some extent (dependent on the fineness of the mesh); a concrete
slab that is cracked in longitudinal tension can be modelled with anisotropic properties; effective
moments, shears and axial forces on composite beam sections can be determined (by the software) from
the stresses determined in the global analysis.
Verification of buckling resistance of members sometimes requires an elastic buckling analysis of the
structure to determine its critical loading. Software is available that can determine elastic buckling load
using finite element models, and these models can be used either to determine elastic critical moments of
beams directly or with the general method of verification (see appendix A of AS 5100.6 for details).
58
In some cases, second-order (large displacement) analysis is also required. This requires more complex
software; modelling and interpretation of the output requires previous experience in this type of analysis.
The use of a 3D finite element model for multi-girder decks allows intermediate bracing to be modelled
realistically, rather than the representative beam elements that need to be included in a 2D grillage in
order to derive the local restraint forces referred to above.
Finite element modelling does have its limitations. These are:
Depending on the level of complexity of the model, the quantity of computation and time required to
analyse the model may be long. This may result in a delay in the availability of the model output,
leading to later confirmation of the design details.
Modelling of concrete especially when cracked is difficult.
Greater experience is needed to construct the model and interpret the results.
6.2.4
Alternative analysis options
In addition to the different form of grillage analysis and finite element modelling outlined above, there are
other computer software packages that provide similar, possibly less detailed, results. The main difference
can be the different modelling engine that the program is based on, its available function and output
capabilities. Depending on the level of sophistication that is required or afforded by the engineer, the
following options can be considered.
6.2.4.1
Simple design tools
These can either be in-house Excel spread sheets or a specially developed program that performs
repetitive or specific functions. Most, if not all, design practices have at least one form of a simple design
tool that are especially useful in the preliminary design stage to assist in evaluating the different bridge
concepts. Once the final concept has been chosen, then the more detailed grillage analysis or finite
element modelling is used.
6.2.4.2
Computer-aided engineering tools
Computer-aided engineering tools (often referred to as CAE) are more sophisticated than simple design
tools that provide a wide range of functions to bridge designers. The functions may include analysis,
simulation, design and detailing - the CAE can import and export to other programs by the use of building
information modelling (BIM). They require additional more detailed inputs than simple design tools, which
in turn results in comprehensive outputs that may include detailed design calculations to a specified
standard to checking creep and cracking in concrete slabs and even producing construction drawings.
These programs mostly use a finite element analysis engine. However, the main difference from finite
element modelling outlined above is that each designed component is taken as a whole or as a very rough
mesh (for slabs) in comparison to the more detailed and finer finite element modelling. This results in a
faster calculation of the results and is easier to set up the model and analyse, it also does not require a
highly specified computer or a specially trained engineer to use the program.
6.3
Deck slab analysis
The Bridge manual refers to NZS 3101 (SNZ 2006) when analysing the concrete deck slab. Two methods
exist: the empirical method, based on assumed membrane action; and elastic plate bending analysis
method.
59
The empirical design method takes account of membrane action in the slab, and is based on test results.
Requirements for the use of the empirical method are given in clause 12.8.2.1 of NZS 3101, where one of
the requirements specifically state that the deck must be fully cast in place. However, as demonstrated in
section 3.5, partial depth precast decking provides a more economical option than cast in place decks. In
this case, elastic plate bending analysis should be used.
Clause 12.8.3 of NZS 3101 outlines the design basis when using the elastic plate bending analysis
method. It should be noted that any of the following methods can be used in this case; simple analysis,
grillage analysis with a fine mesh and finite element analysis. Clause 12.4 of NZS 3101 outlines the design
methodologies that can be used for analysing bridge deck slabs, and also provides guidance on some of
the design options given above. Further details on the design of slabs are given in section 7.10.
The choice of any of the above analysis options is dependent on the level of complexity of the slab design.
For example, a simple analysis may be sufficient for a single span, simply supported bridge, while a grillage
or finite element analysis of the deck may be required for a continuous, multi span, multi-girder bridge. This
is also dependant on the level of experience of the bridge designer and the computer modelling programme
available. As mentioned in section 6.3, it is always recommended that the designer checks the results, for
example, for deck slabs one way is by the use of Pucher Charts (Pucher 1964).
Pucher Charts are a series of contour plots of influence surfaces, as illustrated in figure 6.3. The
simplification of support conditions to permit use of standard charts normally leads to a conservative
assessment of worst moments. If a fine mesh of shell elements is used in a 3D global model, local
bending moments in the slab will be determined directly. A resolution (node spacing) of about 500mm
should be adequate in view of the loaded area under a wheel and dispersal through surfacing and slab.
Figure 6.3
Typical influence chart of slab moments using Pucher Charts
Negative contour
Positive contour
Localised dead load effects including self-weight of the slab must also be considered. However, these
effects should be able to be simply calculated using manual methods. As deck slabs are usually relatively
thin, these effects are normally relatively small.
6.4
Design loading and load combinations
All design loading acting on the bridge structure should be in accordance with section 3 of the Bridge
manual and in accordance with the load combinations given in section 3.5 of that document. These
loadings are summarised below.
60
6.4.1
Traffic loads gravity effects
Rules for the placement of these loadings, number of load lanes per bridge width and other required
details while designing the bridge structure are given in section 3.2.3 of the Bridge manual. Section 3.2.4
provides guidance on the combination of the normal (HN) and overload (HO) loading components and the
reduction factors used when there is more than one element of loading (ie more than one notional lane
loaded).
For all above ground level components of the bridge, the applied HN and HO loading must be multiplied
by the dynamic load factor in accordance with section 3.2.5 of the Bridge manual. For below ground level
components of the bridge, the dynamic load factor taken is 1.0, to allow for the fact that vibration is
damped out by the soil.
Figure 6.4
Diagrammatical representation of normal and overload loading
HN Load Element
HO Load Element
5000
120kN/Axle
5000
UDL 10.5kN/m
120kN/Axle
240kN/Axle
UDL 3.5kN/m
Contact Areas
3000x200
3000
Contact Areas
500x200
240kN/Axle
3000
UDL 10.5kN/m
UDL 3.5kN/m
Alternative (b)
6.4.2
Fatigue loading spectrum
The design loading used in the fatigue assessment needs to represent the expected frequent loading over
the design life of the structure. It is common to represent the varied spectrum of real traffic by a single
notional vehicle that traverses the length of the bridge a defined number of times over the life of the
bridge.
Section 3.2.6 of the Bridge manual currently refers to the Recommended draft fatigue design criteria for
bridges (Clifton 2007a), which is reproduced below. The background to this is given in appendix D of this
guide.
6.4.2.1
Basic loadings for fatigue design
There are two basic loads given, in clause 6.9 of AS 5100.2 (SA 2004), being:
1
The modified individual A160 heavy axle load from AS 5100.2 as shown in figure 6.5.
3000
2100
UDL 3.5kN/m
450
Contact Areas
900x600
450
Alternative (a)
61
Figure 6.5
160 fatigue loading
160 kN
Elevation
1800 mm
400 mm
3000 mm standard
design lane
250 mm
Plan
The modified individual M1600 moving traffic load from AS 5100.2, without the UDL component, as
shown in figure 6.6.

Figure 6.6
M1600 fatigue load

360 kN
360 kN
360 kN
360 kN
Elevation
3750
12501250
Varies 6250 min.
12501250
5000
12501250
1800 m m
400 mm
12501250
3000 mm standard
design lane
250 mm
Plan
6.4.2.2
Fatigue design loading and number of fatigue stress cycles
It is recommended that the fatigue design traffic load effects be determined from 70% of the effects due to
the passage of either a single A160 axle or a M1600 load, whichever gives the more severe range of
effects at the location being assessed. In each case, a load factor of 1.0 is used and the load effects are
increased by the dynamic load allowance (), as given below:
for the A160 axle load
= 0.4
for the M1600 axle load
= 0.3.
Each load should be placed within the width of any traffic lane as marked on the bridge such that the
fatigue effects for the component under consideration are maximised.
Unless determined otherwise by the relevant authority, the number of fatigue stress cycles to be used for
the calculation of the fatigue capacity of the structural element under consideration should be as follows:
1
For the fatigue design load of 0.70 x (A160 axle load) x (1 + ):

(current number of heavy vehicles per lane per day) x 4 x 104 x (route factor).
For the fatigue design load of 0.70 x (M1600 moving traffic load without UDL) x (1 + ):
(current number of heavy vehicles per lane per day) x 2 x 104 (L-0.5) x (route factor).
In the above expression:
62
The current number of heavy vehicles per lane per day is provided in the bridge project
documentation or is determined by project specific traffic counts. The current number of heavy
vehicles is based on the year the bridge is to be put into service.
On rural routes where there are two or more lanes in one direction, the number of heavy vehicles per
lane per day shall be the total of the heavy vehicles travelling in that direction. On urban routes where
there are two or more lanes in one direction, the number of heavy vehicles per lane per day shall be
65% of the total number of heavy vehicles in that direction.
The route factor for the class of road shall, unless specified otherwise by the relevant authority, be as
follows:
for principal state highways
= 0.9
for major urban roads
= 0.65
for other rural routes
= 0.45
for urban roads other than major urban roads
= 0.3
L is the effective span in metres and is defined as follows:
for positive bending moments, L is the actual span in which the bending moment is being
considered
for negative moment over interior supports, L is the average of the adjacent spans
for end shear, L is the actual span
for reactions, L is the sum of the adjacent spans
for cross-girders, L is twice the longitudinal spacing of the cross-girders.
AS 5100.2 states that the fatigue design traffic load effects and relevant stress cycles should be applied to
each design lane independently. However, current New Zealand practice is to apply this load and relevant
stress cycles on each marked lane independently. The rationale being that in reality the traffic load will be
acting on the actual marked lane rather than on the hypothetical design lane. Therefore, the number of
marked lanes given in the project design statement should be designed for; this includes any future
marked lanes that will be added to the bridge structure.
The fatigue stress range shall be taken to be the maximum peak-to-peak stress from the passage of the
relevant fatigue design load.
These recommendations do not apply to fatigue design of roadway expansion joints.
6.4.3
Traffic loads - horizontal effects
Section 3.3 of the Bridge manual gives the required forces for determining the horizontal effects of traffic
loadings, namely braking and traction in section 3.3.1 and centrifugal force in section 3.3.2.
6.4.4
Loads other than traffic
Table 6.1 lists the other non-traffic loads given in section 3.4 of the Bridge manual that must be
considered when designing a bridge.
63
Table 6.1
Other non-traffic load cases
Load
Section
Dead
3.4.1
Superimposed dead
3.4.2
Earthquake
3.4.3
Shrinkage, creep and
3.4.4
Note
Given as 1.5kN/m2 for road surfacing, in addition to handrails, guardrails,

lamp standards, kerbs and services.
prestressing effects
Outlines the effects of shrinkage and creep of concrete, and shortening due
to pre-stressing (see section 6.4.5).
Wind
3.4.5
Temperature effects
3.4.6
Construction and
3.4.7
See section 6.4.5 for further details.
maintenance
Water pressure
3.4.8
Groundwater on buried
3.4.9
Outlines the water pressure loading experienced by the bridge piers in

waterways.
surfaces
Water ponding
3.4.10
Snow
3.4.11
Earth
3.4.12
Outlines the snow loading requirements that are determined from AS/NZS
1170.3 (SNZ 2003a). Note this only applied to pedestrian bridges.

Outlines the provisions for determining the earth loads experienced by the
bridge substructure.
Kerbs, guardrails, barriers
3.4.13
Footpaths and cycle
They should be designed in accordance with appendix B of the Bridge
manual.
and handrails
3.4.14
These are designed to the loads given in section 3.2 of the Bridge manual.
tracks
Vibration
3.4.15
Settlement, subsidence
3.4.16
and ground deformation

Forces locked-in by the
The most adverse combination of differential ground movement should be

considered. Advice should be sought from the geotechnical engineers.
3.4.17
The guidance given in clause 4.6 of AS 5100.6 should be followed.
erection sequence
Collision
6.4.5
3.4.18
Shortening, temperature effects and ground deformation
Section 3.4.6 of the Bridge manual outlines two types of temperature effects. The first is the overall
temperature change which is taken as 25C for steel and is the value that is used for composite sections.
This covers the forces and movements resulting from the variation in the mean temperature of the
structure.
The second is the differential temperature change which for non-compact sections is applied at both the
ULS and SLS load cases, but only for the SLS case for compact sections. This is due to the compact section
being able to redistribute the forces due to the nature of the compactness of the section. This covers the
allowable stresses both longitudinal and transverse of the section resulting from the temperature variation
through the depth of the structure.
64
Similar to differential temperature, the effects of shortening and settlement, subsidence and ground
deformation should be applied to non-compact sections at both the ULS and SLS load cases, and only for
the SLS case for compact sections.
6.4.6
Geotechnical actions in integral bridges
Geotechnical actions on bridge superstructures are the pressures of the backfill at the abutments. These
pressures arise initially from the backfilling behind the abutments but increase over time due to the strain
ratchetting effect from the cyclic expansion/contraction of the deck with temperature variation. Soil
pressures due to the movements at the ends of the deck may be calculated using guidance given in
section 4.8 in the Bridge manual and additional information may be found in PD 6694-1 (BSI 2011).
6.4.7
Load combinations
Section 3.5 of the Bridge manual outlines the parameters and load combination of the load effects in
section 3.4 of the Bridge manual, for both SLS and ULS.
6.5
Seismic effects
Section 5 of the Bridge manual is used to determine the horizontal force generated by the seismic action
on the bridge superstructure. These forces in turn are transferred into the supports (piers and abutments),
which can be designed to form plastic hinges that resist this horizontal force (as shown in figure 5.3 of
the Bridge manual). Designers should be aware that the horizontal forces are transferred by the restraints
into the supports, which are designed for; this is discussed further in section 7.6.7. Therefore, seismic
actions acting on the superstructure are actually minimal as long as the load path is available to transfer
these actions into the substructure.
The design of the piers and abutments is outside the scope of this guide.
65
Detailed design
Once the final bridge configuration has been chosen and the preliminary design completed, the designer
then has to refine and verify the performance of the main structural components and to detail the
connections. The main design steps include:
The resistance (capacity) of the girders and other structural elements must be verified at the ULS
during construction and in service. This requires consideration of the bending resistance of the steel
and composite cross sections, the shear resistance of the web panels and the shear connection
between the steel and the concrete slab.
The resistance to buckling needs to be verified, for the bare steel girders during construction and, for
the composite girders, for the bottom flanges of composite girders adjacent to intermediate supports,
where they are in compression.
In integral bridges, the effect of coexisting compressive forces must be taken into account.
The resistance of the deck slab to combined local and global effects must be verified.
Adequacy at the fatigue limit state must be verified.
Toughness of the steel against brittle fracture must be ensured by selecting a suitable material subgrade.
Performance at the SLS requires to be verified; principally the checks are to ensure that inelastic behaviour
does not occur under SLS actions, that crack widths in concrete are not excessive and that deformations
are within acceptable limits.
The following sections outline the design procedures to conduct a detailed design of the above
components, by providing a design flow chart for each component, while referencing the appropriate
Clauses in AS 5100.6 (SA 2004) and the corresponding sections in this document. All clause references are
to AS 5100.6, unless stated otherwise. Figure 7.1 outlines the overall design procedure that should be
undertaken when designing a steel-concrete composite bridge.
Figure 7.1
Overall design flow chart for the design of a steel-concrete composite bridge
Loading cases
Determine appropriate cases
for maximum effects
Section 6.4
Section properties
Global analysis
Carry out elastic analysis and member
buckling analysis of bridge model
Calculate the member section

properties at critical section(s)
Section 7.1.1
Clauses 4.2 and 4.3/section 6

Checks at ultimate limit state
Checks at serviceability limit state
Main girders
Compact and non-compact beams
Section 7.1.2
Section 7.5
Restraints
Section 7.6
Longitudinal shear force
Section 7.7
Longitudinal shear connection
Connections
Section 7.7
Section 7.8
Connections
Reinforced concrete deck slab
Section 7.8
Section 7.10
Web stiffeners
Section 7.9
Reinforced concrete deck slab
Section 7.10
66
Detailed design
7.1
Overall bridge design
7.1.1
Section properties
The section properties of the main members should be determined at key locations. Section properties are
required for the bare steel members in the construction stages and for the composite members at the inservice stage. The main section properties required are:
area of the cross section, A
second moment of area about the major axis, Ix
plastic section modulus, S
elastic section modulus at extreme fibres, Z
torsion stiffness, J
out of plane stiffness, Iyy.
For composite members, two sets of properties are needed: short-term properties for effects due to live
loads, including wind, temperature and seismic loading; long-term properties, for superimposed dead
loads and shrinkage.
The short-term and long-term properties are usually calculated in steel units by using the transformed
area of the concrete slab dividing the area by the modular ratio (n), the ratio of the elastic modulus of
steel (Es) to the modulus of elasticity of concrete (Ec). For the short term loading the normal modulus of
elasticity of concrete is used, while for long term properties the effect of creep is taken into account,
which is given by the creep factor ( ) given in clause 5.2.1 of NZS 3101 (SNZ 2006).
cc
The modulus of elasticity for steel and for concrete is given in clause 2.2.5 of AS 5100.6 (SA 2004) and
clause 5.2.3 of NZS 3101 respectively. The modular ratio (n) values for short and long-term loading
derived for typical strength grades of concrete are given in table 7.1, for concrete with a density () of
2400kg/m3 (normal weight concrete).
Table 7.1
Modulus of elasticity for steel and concrete
Modular ratio (n)
Standard strength grades of concrete (MPa)

25
32
40
50
65
Short-term loading
8.5
7.8
7.2
6.6
5.9
Long-term loading
44.3
34.3
25.1
19.8
17.8
When the steel girder is acting compositely with the concrete deck slab, the effect of shear lag in the
concrete deck slab should be taken into account, in accordance with clause 6.1.7. To calculate the
transformed area of concrete, the effective width of the concrete slab is required, which is given in clause
4.4.1 and is represented in figure 7.2.
67
Figure 7.2
Effective width of concrete
beff
beff
tslab Girder 1: Composite flanges on both sides of the web:
beff = min(Le/5, Lg, 12*tslab)
Girder 2: Composite flanges on one side only of the web:
Lg
beff = min(Le/10, Lg/2, 6*tslab)
Girder 2
Girder 1
Where:
beff is the effective width of the concrete slab.
Le
is the effective span length of the girder, taken as distance between supports for simply supported
beams and girders, and 0.7 times the distance between supports for continuous beams and girders.
Lg
is the distance centre-to-centre of girders.
tslab is the the least thickness of the slab.

The same effective width is used over the whole span, for both hogging and sagging regions (except for
the determination of fatigue stresses in hogging regions in continuous girders, where the effective width
values are taken as half, see further comment in section 7.10). However, when there is a difference in the
decking thickness in different spans, for example at intermediate supports between unequal spans, then
the effective width based on that spans thickness will be used. The transformed area of concrete is only
taken into account when calculating the properties for the sagging moment region. In this region the
concrete is uncracked and in compression, therefore contributing to the composite action between the
steel girder and the slab, as stated in clause 6.3.1. In the hogging region, the contribution from the
concrete is neglected and instead the contribution from the reinforcing bars is taken into account, as
stated in clause 6.3.2.
Examples of the elastic and plastic stress blocks for both the sagging and hogging regions are shown in
figure 7.3 below. Calculation of plastic blocks is demonstrated in appendix E of AS 5100.6 for sagging
regions.
Figure 7.3
Stress distribution in composite beams in the sagging and hogging moment region
Elastic
Sagging Moment Region
Plastic
Elastic
Hogging Moment Region
Plastic
68
Detailed design
7.1.2
Section slenderness and the definition of compact and non-compact

sections
Cross sections are classified according to the slenderness of the elements of the cross section that are in
compression. Two principal classes are defined, compact and non-compact (actually not compact in AS
5100 but the hyphenated term is used in this guide, for greater clarity). Section slenderness limits for
steel cross sections given in clauses 5.1.2 and table 5.1 are used to classify the section as compact or
non-compact and these limits also apply to composite sections (see clause 6.1.8) with an extra limitation
applied to steel flanges connected to a concrete slab. Compact sections are those where the full plastic
moment can be developed before, and maintained after, the onset of local buckling (clause 5.1.3). The
web and compression flange possess sufficient stiffness to enable full plasticity and adequate inelastic
rotation to be developed without the loss of strength due to local buckling. In a compact section, the
section slenderness ( ) is less than or equal to the plasticity slenderness limit ( ); ie .
s
sp
sp
Clause 5.1.4 states that a section that is non-compact is one for which local buckling prevents the
development of the full plastic moment and which is liable to local buckling before the onset of yielding.
In non-compact sections the section slenderness ( ) is greater than the plasticity slenderness limit ( ), ie
s
s >
.
sp
sp
Attention is drawn to the sub-division of the non-compact classification, effectively creating a class called
slender, although this term is only used explicitly in AS 5100.6. In a non-compact section that is not
slender the section slenderness ( ) is greater than the plasticity slenderness limit ( ) but less than or
s
sp
equal to the yield slenderness limit (), ie < . In a slender section the section slenderness is
sy
sp
sy
greater than the yield slenderness limit, ie > . By combining the non-compact and slender section
s
sy
definitions, AS 5100 has created potential confusion. Note that, generally, it would be uneconomic to use
slender flange elements in a girder, but web elements may quite often be slender.
Be aware that the term slender section should not be confused with slender beam.
Plastic moment capacity is often expressed in relation to the plastic section modulus (S), which is
determined by consideration of rectangular stress blocks. For composite sections, the plastic section
modulus is usually expressed in steel units and the stress block for the concrete slab depends on the
strength of the concrete, relative to the yield stress of the steel.
Elastic moment capacity depends on the stress at the extreme fibre (the furthest from the neutral axis, in
either tension or compression, expressed as the elastic section modulus (Z )e but for summation of stresses
in composite sections (see discussion in section 7.3.2) the section moduli will be needed at several
locations in the cross section. When a cross section is slender, part of the cross section is deemed
ineffective and this reduces the value of the second moment of area and thus the elastic section moduli
are also reduced.
It should also be noted that although AS 5100.6 allows plastic theory for the determination of section
capacity when the section is compact (and this is particularly applicable for composite girders, see clause
6.3.3 and appendix E of AS 5100.6) elastic analysis is still required for determining action affects (clause
4.1), apart from the rare situations where a rigorous analysis of the whole structure is carried out.
7.2
Main girders at ULS
The main longitudinal girders must be designed to provide adequate strength in bending and shear, to
resist the combined effects of global bending, the local effects such as compression over bearings, and
structural participation with the bracing systems. Adequacy must be demonstrated during construction (at
69
the bare steel stage and the partly composite stage, when the deck is concreted in stages) and in service
(when the girder is composite). Generally, the design procedure for both situations is similar but, for
clarity, the sequence is shown separately below for the non-composite bare steel girders and for
composite girders, in figures 7.4 and 7.5 respectively.
Cross girders in ladder deck bridges follow the same sequences; bracing members not acting compositely
with the deck follow the sequence in figure 7.4.
Figure 7.4
Non-composite bare steel girder design flow chart

Determine the section slenderness limit
Section properties
Clause 5.1.2/section 7.1.2
Section 7.1.1
Non-compact section
Compact section
Calculate nominal section moment capacity (M )

s
Clause 5.2/section 7.3.2

Lateral and torsional
Girder with full lateral restraints
Girder without full lateral restraint
restraint
U-frame restraint
Determine the effective length
Clause 5.6.3 or 5.6.5/section 7.3.2

Determine moment modification factor (m)
Reference buckling
Clause 5.6.1 or 5.6.2
bending moment (Mo)
Clause 5.6.1.1
Determine slenderness reduction factors (s)
Elastic buckling bending
Clause 5.6.1 or 5.6.2
moment (Mob) from

Calculate the nominal member moment
elastic buckling analysis
capacity (Mb)
Clause 5.6.6
Calculate vertical shear capacity
Web stiffener spacing (s) for stiffened
Clause 5.10/section 7.4
Redesign
web only
Check for adequacy for bending moment, shear and bending
Design bending moment (M*) and
moment/shear interaction.
design shear force (V*) from
Clause 5.1.6 or 5.1.7, 5.10.1 and 5.11/sections 7.3.1, 7.4 and

7.4.2
global analysis.
No
OK?
Yes
Satisfactory
Clause 4.2
70
Detailed design
Figure 7.5
Composite main girder design flow chart
Determine the section slenderness limit
Section properties
Section 7.1.1
Compact section
Non-compact section
Calculate nominal section moment capacity (Ms)
Girders continuously restrained by
Girders continuously restrained by deck
deck at compression flange level
not at compression flange level
Nominal member
moment capacity (Mb) of
the bare steel section
Clause 5.6
Lateral torsional buckling
Clause 6.4.2.1/section 7.3.2
Lateral distortional buckling

Calculate the nominal member moment
capacity (Mb)
Calculate vertical shear capacity
Web stiffener spacing (s) for
Clause 6.5/section 7.4
stiffened web only
Check for adequacy for bending moment, vertical and

longitudinal shear, and bending moment/shear interaction.
Clauses 6.2.1 or 6.2.2, 5.10.1 and 6.6/sections 7.3.1, 7.4 and

7.4.2
Design bending moment

(M*) and design shear force
(V*) from global analysis. Or
summation of stresses for
non-compact sections for
bending check. Clause 4.2 or
4.6
Redesign
No
7.3
Design for bending
7.3.1
Bending capacity
7.3.1.1
Compact sections
OK?
Yes
Satisfactory
For a member with a compact cross section bent about the major (x-x) axis, the requirement for bending
capacity is expressed, for steel sections, in clause 5.1.6 as:
(Equation 7.1)
(Equation 7.2)
For composite sections, the requirement is expressed in clause 6.2.1 as:
(Equation 7.3)
(Equation 7.4)
71
The capacity reduction factor () is given in table 3.2 of AS 5100.6 (SA 2004).
For both sets of requirements, the right-hand side of the first expression represents the bending capacity
of the cross section and the right-hand side of the second represents the bending capacity of the beam,
although those terms are not actually used in AS 5100.6.
Note that, since the plastic moment can be developed and maintained in a compact section, the
requirement is simply that the total design bending moment (M* x) shall not exceed the bending capacity at
ULS; there is no need to check the summation of the separate ULS stress distributions for each stage of
construction.
Also, steel beams that are compact when acting compositely with the slab may not be compact when
acting alone during construction. In such a case, the checks for the construction condition must be made
on the basis of non-compact sections, as stated in clause 6.1.2.
Another point of interest is the difference in the acting design moments at the sagging and hogging
moment regions, which are typically located in the mid-span and intermediate supports respectively. This
may result in different section classifications for those regions, ie the girder may be compact in the
sagging region and non-compact in the hogging region. This is especially important for simply supported
bridges that may use the same section in both regions. In this instance, the girder as a whole (in both
sagging and hogging moment regions) should be treated as a non-compact member and summation of
stresses apply.
In most cases of composite beam and slab construction the hogging moment regions over intermediate
supports are unlikely to be compact. To achieve compact classification, the web would need to be much
thicker than what is required for shear capacity and therefore would be uneconomic.
7.3.1.2
Non-compact sections
For a member with a non-compact cross section bent about the major (x-x) axis, the requirement for
bending capacity is expressed, for steel sections, in clause 5.1.7 as:

(Equation 7.5)
(Equation 7.6)
For composite sections, the requirement is expressed in 6.2.2 as:

(Equation 7.7)
(Equation 7.8)

(Equation 7.9)
0.62
The capacity reduction factor () is given in table 3.2 of AS 5100 (SA 2004).
For both sets of requirements, the right-hand side expressions represents the design value of the
material strength, as limited by buckling bending capacity of the beam, although the term design value is
not used in AS 5100.6.
The design capacity of non-compact beams in bending is determined by elastic stress distributions and
limiting stresses and in most cases checks need only be made at ULS (since the design bending moments
at SLS will usually be lower). Checks at SLS are required in only a few circumstances, such as when
checking the effects of shrinkage and differential temperature (see section 7.5.3 for further details).
A non-compact section is not able to redistribute stresses once yield is reached in the compression
flange and therefore the verification must be made on the summation of the elastically determined
stresses at each construction stage.
72
Detailed design
The total stresses and strains in the fibres of a composite beam where the deck slab is added in stages are
determined as the summation of the distributions for each stage (and similarly for short and long-term
loads), as shown diagrammatically in figure 7.6. The position of zero stress at ULS will therefore not
necessarily correspond with any particular neutral axis level.
Figure 7.6
Summation of stresses for staged construction
Steel
Section
Long-term
Composite
Section
Short-term
Composite
Section
Total Stress
Distribution
Where SLS must be checked as well as ULS, the stress distributions for SLS and ULS must be calculated
separately, each using its appropriate set of partial factors for the various loads.
7.3.2
Section moment capacity
7.3.2.1 Compact sections

For steel compact sections, the nominal section moment capacity (Ms) is given by clause 5.2.1 as:
=
(Equation 7.10)
Where the effective section modulus (Z e), is equal to the lesser of the plastic section modulus (S) and
1.5 times the elastic section modulus (Z) (clause 5.1.3), ie:
Ze = S 1.5Z
(Equation 7.11)
For composite compact cross sections the nominal section moment capacity (Ms) is given by clause 6.3.3
as:
= (Equation 7.12)
Guidance on determining (Mp) is given in appendix E of AS 5100.6 (2004) for sagging moment regions.
7.3.2.2
Non-compact sections
In the verification criteria, reference is made only to stresses; although the section moment capacity (Ms) is
defined in clauses 5.2.2 and 6.3.4 and reference to it is only made in relation to interaction between
bending and shear (see section 7.4.2).
For a steel non-compact section, the nominal section moment capacity (Ms) is given by clause 5.2.2 as:
=
(Equation 7.13)
The designer should be aware that the effective section modulus (Ze) is the elastic section modulus of the
effective section; this in turn is affected by the section classification. As discussed in section 7.1.2,
AS 5100.6 combined the non-compact and slender classifications and the calculation of the effective
section modulus is not clear. Based on the guidance given in AS 5100.6 and Gorenc et al (2005), the
effects of the section classification on the effective section modulus, are summarised below:
73
For non-compact sections, where sp < s sy (clause 5.1.4), the effective section modulus (Zen) is
between the plastic section modulus (S) and the elastic section modulus (Z). This can be calculated by
using;
Zen = Z + cz (Zc Z)
(Equation 7.14)
Where:
Zc is the effective section modulus assuming the section is compact (ie given by equation 7.11)
cz is the
For slender sections, where s > sy (clause 5.1.4), the section is less effective, therefore the effective
section modulus (Zen) is less than the elastic section modulus (Z). There are two situations that should
be considered in bridges, these are:
section elements having uniform compression (ie no stress gradient), such as flanges of a UB bent
about the major axis.
method 1: (simple method)

sy
Z =Z
en
method 2: (clause 5.1.4)
(Equation 7.15)
The effective section modulus is calculated for the effective cross section determined by
removing the excess width of plates whose (b/t) exceeds the ey for that element. See clause
5.1.5 for further information.
sections with slenderness determined by a stress gradient in plate elements with one edge
unsupported in compression, such as a UB bent about its minor axis.
sy
7.3.2.3
Composite stage
Zen = Z
(Equation 7.16)
Clause 6.3 covers the calculation of the nominal section capacity in the composite stage. The main
difference is when determining the effective section modulus (Ze), the concrete deck is taken into account.
As explained in section 7.1.1 and demonstrated in figure 7.3, the transformed area of the concrete is
taken in the sagging moment region, while in the hogging moment region only the contribution from the
steel reinforcement is taken into account.
7.3.3
Member moment capacity
The moment capacity of a member, whether the girder is non-composite or composite, depends on both
the moment capacity of the cross section and on the degree of lateral restraint provided to the critical
flange, which for a member supported at both ends is the flange in bending-induced compression.
A steel bridge beam in bending is required to be provided with effective lateral and torsional restraint
(about its longitudinal axis) at its supports. Between supports it may be unrestrained, especially during the
construction stage. For such a configuration, the member can buckle in a lateral-torsional buckling mode
(see section 7.3.4) and the moment capacity of the member depends on both the section moment capacity
and the elastic critical buckling moment (referred to in AS 5100.6 as the reference buckling moment M ).
o
The value of the elastic critical moment can be increased by providing intermediate restraints between the
supports; the beam can then be considered as a series of segments, each with full or partial restraint at
74
Detailed design
the ends of the segments. In a composite bridge, the presence of the deck slab provides continuous
restraint to the top flanges, but, adjacent to intermediate supports, it is the bottom flange that is in
compression. Therefore, buckling of the segments between the bracing positions to the cross section for
that bottom flange must be considered.
The forms of restraint considered by AS 5100.6 are discussed in section 7.3.6 below. The type and
spacing of the restraints determines the member moment capacity, as discussed below. However, one
form of buckling for bare steel girders is not recognized by AS 5100.6; this is discussed separately in
section 7.3.7.
7.3.3.1
Segments with full lateral restraint
Where the ends of a segment of steel beam are fully or partially restrained and the length is sufficiently
short that the slenderness (expressed in terms of the ratio L/ry, where L is the length of the segment and
r is the radius of gyration of the beam about the minor axis) does not exceed a limiting value, the
y
segment is considered to have full lateral restraint. Clause 5.3.1 states that, with full lateral restraint, the
nominal member moment capacity (M b) is taken as equal to the nominal section moment capacity (M ). The
s
slenderness limits are given in clause 5.3.2.4.

For a composite beam in sagging, the compression flange is continuously restrained and, again, M = M
b
(see clause 6.4.1).

7.3.3.2
Segments without full lateral restraint
When the slenderness exceeds the relevant limit for considering the segment as having full lateral
restraint, a slenderness reduction factor needs to be applied to the section moment capacity to determine
the member moment capacity. The value of the reduction factor and the means for its determination
depend on the type of cross section and type of end restraint. The rules for the various situations covered
in clause 5.6.1 are discussed below.
Segments fully or partially restrained at both ends

For a non-composite girder with open sections with equal flanges, the nominal member moment capacity
(Mb ) is given by clause 5.6.1.1(a) as:
= (Equation 7.17)
Where is a moment modification factor and s is the slenderness reduction factor, which depends on
m
the elastic buckling moment (M ) (either given by clause 5.6.1.1 or determined by an elastic buckling
oa
analysis).
For non-composite girders with open I-sections with unequal flanges, the nominal member moment
capacity (Mb) is given by clause 5.6.1.2, which uses the same expression as clause 5.6.1.1(a). However, to
take into account the unequal flanges a different reference buckling moment (Mo) expression is used. Both
cases use the value of effective length given by clause 5.6.5.
For the support regions of continuous composite beams, clause 6.4.2.2 does provide rules for
determining a reduction factor for distortional buckling but this relies either on a buckling analysis of the
structure or on a simplified evaluation that considers only the whole span and does not take any account
of bracing near the support. Although clause 6.4.2.1 applies to situations where there are discrete U-frame
restraints, it is not appropriate for the hogging moment regions (the evaluation of effective length in
clause 5.6.4 has been derived for situations where the moment is relatively constant).
75
Segments unrestrained at one end (ie cantilever segments)

For a non-composite girder with a segment unrestrained at one end, the nominal member moment
capacity (Mb) is given by clause 5.6.2, again using the value of effective length given by clause 5.6.5.
Beams restrained by U-frames

For non-composite girders restrained by U-frames, the nominal member moment capacity (Mb) is again
given by clause 5.6.1.1(a) but the effective length is now given by clause 5.6.3.
7.3.3.3
Effective length
The different calculations for effective length are summarised below:
Clause 5.6.3: Effective length for beams restrained by U-frames

U-frame restraints are created by stiff transverse beams, connected close to the tension flange, acting in
conjunction with either plan bracing or deck slab at tension flange level, and with web stiffeners on the
webs of the beams. Intermediate U-frames are designed in accordance with clause 8.4.6
For a ladder deck bridge to provide restraint to the bottom compression flange (in the hogging moment
region), a moment connection is required between cross-beam and stiffener in order for this connection to
provide P or F cross section restraint.
Clause 5.6.4: Effective length for beams continuously restrained by a deck not at compression flange level
This clause is mainly applicable to half-through deck bridges, and is outside the scope of this document.
Clause 5.6.5: Effective length (of a segment)

This clause gives rules for the effective length of segments and sub-segments (between intermediate
restraints) for full partial and unrestrained conditions, as defined by clause 5.4 (see section 7.3.6).
Allowance for restraint of the compression flange in plan at the ends is included, though such restraint is
not normally present.
7.3.4
Lateral buckling
The buckling of a length of beam without intermediate restraint under bending is generally referred to as
lateral-torsional buckling3.
Lateral-torsional buckling of a beam occurs when a beam is loaded in flexure (bending) and the critical
flange (which is the compression flange) is not restrained in the lateral direction (ie perpendicular to the
plane of bending) over a length between positions of effective cross-section restraint. At a certain critical
limit, the beam will fail in a mode where the critical flange displaces laterally and the beam twists, as
shown in figure 7.7.
Lateral-torsional buckling is a term that is widely used internationally and is the term for this form of instability in the
Eurocodes. AS 5100.6 (SA 2004) does not use the term, although it does refers to elastic flexural-torsional buckling
analysis in clause 5.6.6. The term flexural-torsional buckling is used in the Eurocodes for a form of instability in
members subject to compression. Three types of instability are possible for members in compression: flexural buckling
(the most common type of buckling), torsional buckling (involving twist but no lateral displacement) and flexuraltorsional buckling (a combined mode). However the first two of these relate to members subject to axial compression
which is not the case for bridge beams.
76
Figure 7.7
Detailed design
Lateral-torsional buckling of an I-girders critical flange between adjacent points of restraint
The critical load at which the beam buckles may be increased by providing intermediate restraints and its
value then depends on the types of restraints and their strength and stiffness. If torsional restraint at the
support is not effectively rigid, the critical load will be reduced; this too must be taken into account.
A composite continuous beam (with the slab connected to the top flange) may also be limited by buckling,
when the bottom flange buckles laterally adjacent to an intermediate support. This is actually a
distortional buckling mode but is conservatively treated in the same manner as lateral-torsional buckling;
the critical load depends on the beam geometry and any restraints to the bottom flange.
7.3.5
Restraints
A restraint system (bracing) is required at the supports of the main girders to provide a load path for
transferring horizontal forces (transverse to the main girders) to the bridge supports and to restrain the
main girder against lateral deflection or twist rotation at different stages during construction of the bridge
and throughout its design life. Such restraints are designed in accordance with clause 8.4, which is
discussed in section 7.6 below.
Restraints are classified in clause 5.4 according to the degree of restraint that they provide to the cross
section and the critical flange. The critical flange is the flange that would deflect the furthest laterally under
lateral-torsional buckling (clause 5.5.1). For members with both ends supported (ie not cantilevers) the
critical flange is the bottom flange, which is in compression. For a cantilever which is only restrained at the
support, the critical flange is the top flange (which is in tension). Cantilevers which are restrained against
cross section twist at the unsupported end will then have the critical flange become the bottom flange, which
is in compression, which is why both flanges in such a cantilever are considered critical in design.
7.3.6
Types of restraints
Clause 5.4 outlines the different restraint types and the conditions they must satisfy. The different types
of restraints and other useful terms are summarised below.
7.3.6.1
Fully restrained (F)
There are three cases where a cross section can be considered fully restrained:
A restraint that prevents the lateral displacement of the critical flange of the cross section and
prevents twisting of the section. This case occurs at the support of the main girder provided that the
restraint is designed to resist the appropriate restraining forces from section 7.3.7.
77
A restraint that prevents lateral displacement of the critical flange of the cross section and partially
prevents twisting of the section. This case occurs in the sagging moment region (top flange critical)
and where the restraint is designed to resist the appropriate restraining forces from section 7.3.7.
A restraint that prevents lateral displacement at a point of the cross section other than the critical
flange and prevents twisting of the section. This case occurs in the hogging moment region (bottom
flange critical) when the deck is in place and where the web at the cross section has a full height
transverse stiffener.
7.3.6.2
Partially restrained (P)
There are two cases where a cross section can be considered to be partially restrained.
1
A restraint that prevents lateral displacement at a point of the cross section other than the critical
flange and partially prevents twisting of the section. This case occurs in the hogging moment region
when the deck is in place and where the web is unstiffened, provided that the unstiffened web can
resist the restraining force. For many bridge beams this will not be the case. Guidance on how to
determine the adequacy of an unstiffened web is given in Clifton (1997).
A restraint that prevents twist rotation of the cross section and provides partial restraint against
lateral deflection of the cross section. This case applies during construction when neither the deck nor
any horizontal bracing system is in place and the cross section is stiffened with a moment connection
to the cross beams and both the connection and the cross beams designed to the restraining forces
from section 7.6.
7.3.6.3
Rotationally restrained (R)
A restraint that prevents rotation of the critical flange in its plane. This condition is not usually achieved
and it is conservative to ignore rotational restraint
7.3.6.4
Laterally restrained (L)
The restraint effectively prevents lateral displacement of the critical flange of the cross section without
preventing twisting of the section. This is found in the sagging moment region, where the top flanges are
tied together by the web members of a horizontal truss used for stability during erection.
7.3.6.5
Unrestrained (U)
This is a cross section that does not comply with types fully (F), partially (P) or laterally (L) restrained, ie
does not prevent lateral displacement or twisting of the section.
7.3.6.6
Segment
The length of a beam between adjacent points of full or partial cross section restraint, (F or P) or between
a full or partial cross section restraint and an unrestrained end in the case of a cantilever. The cantilever
case is rare in bridge girders. For a beam having FF or PP end restraints and no mid-span restraints, the
segment length is equal to the beam span. An additional lateral restraint at mid-span would result in a
sub-segment length of one-half span with end restraints of FL or PL.
7.3.6.7
Sub-segment
A segment can be subdivided into portions having at least the lateral (L) restraints to the critical flange at
their ends. Restraint combinations can be FL, PL or LL. Note that sub-segments are used in AS 5100.6 (SA
2004) and this is the notation used in this guide. This is different from NZS 3404 (SNZ 1997b) where the
sub-segment concept has been deleted. It is an unnecessary complication but for consistency of notation
with AS 5100.6 is retained in this guide. It also has more relevance to bridges as pairs of beams restrained
only by cross beams cannot be considered laterally restrained even when the cross beams are located near
78
Detailed design
the critical flange. They have to be designed for the restraining moment and hence provide partial
restraint.
Figure 7.8 shows the type of restraint offered to the girder and the potential location of the critical flange
in different restrained girders. Note that the type of restraint in some cases is dependent on which flange
is critical, ie whether the bending moment at that point along the beam is hogging or sagging. Further
details on restraints and their design are given in section 7.6.
Figure 7.8
Examples of restraining systems for prevention of lateral-torsional buckling failure of simply
supported beams
C = critical flange
= flange to be restrained
C
or C here
or C here
Full or Partial Restraint
Continuously restrained by deck
Deck not at compression

flange level
Lateral Restraint
Partial Restraint
One form of restraint to bare steel beams that is not covered by AS 5100.6 is the intermediate torsional
restraint (without any lateral restraint), shown in figure 7.9. This is achieved by pairing beams together
with transverse planes of triangulated bracing or by stiff cross beams that are rigidly connected to web
stiffeners on the main girders.
Triangulated bracing is often used in multi-girder construction (see section 2.1.3) though cross girders are
used when the main girders are relatively shallow (up to about 1200mm deep). Cross beams are usually
lapped and bolted to the web stiffeners; often the ends of the beams have deep gusset plates welded to
them so that the bolt group has a higher moment capacity. Once the deck slab has been cast, torsional
bracing effectively becomes part of a lateral bracing system.
The ladder deck form of construction, in which the cross beams are rigidly connected to stiffeners on the
main girders, provides a series of closely spaced torsional restraints. Once the deck slab has been cast the
cross girders effectively form inverted U-frames and these may be used to provide some restraint to the
bottom flanges adjacent to intermediate supports. An alternative design for lateral restraint to that offered
by AS 5100.6 has been developed in the UK for ladder deck bridges. Design provisions are given in
PD 6695-2 (BSI 2008) and that guidance is reproduced, in a format compatible with AS 5100.6
terminology, in appendix E.
The torsional flexibility can be evaluated for a unit torque at each beam, in the same sense. There are two
components to the resulting displacement, one due to the vertical deflection of the main girders and one
due to the double curvature flexure of the cross beam. The presence of this torsional restraint will
increase the elastic buckling moment (and thus decrease the slenderness) relative to that for the span
without any intermediate restraint but it is not as effective as plan bracing.
79
Figure 7.9
Determination of torsional flexibility
R2
Twist due to deflection of main girders

Twists in the opposite sense.
R2
R1
Twist due to bending of cross girder

Twists in the same sense.
7.3.7
Design of restraints to AS 5100
Section 7.6 covers the forces required for design of restraints, note that AS 5100.6 (2004) does not
require a stiffness check for restraints that are designed to the specified level of force or moment. This is
the same approach taken in NZS 3404 (SNZ 1997b) and follows from research referenced in Mutton and
Trahair (1975) and Clifton (1994). It is one reason why the restraining force is set at 2.5% of the maximum
critical flange force. Other standards specify a lower level of restraining force, typically 1% in conjunction
with a stiffness requirement.
When the main girders are F or P restrained by cross girders with a moment connection then the
connection and the cross girder must be designed to resist the following:
1
When the bottom flange is critical, the moment is that determined from 2.5% of the moment induced
compression force in the critical flange of the main girder acting at a lever arm corresponding to the
vertical distance between the centroid of the bottom flange and the centroid of the cross girder. For
pairs of main girders so connected and without a deck or plan bracing in place, this provides P
restraint and the displaced shape will be as shown in the top diagram of figure 7.9, involving the two
critical flanges moving away from each other. The major principal axis bending moment induced in
the cross girder is additional to the moment on that girder from the direct loading.
When the top flange is critical, the girders are anchored by the bottom flange which is in tension and so,
for a pair of main girders, the displaced shape will be as shown in the bottom diagram of figure 7.9,
involving the two critical flanges moving in the same direction as they are rigidly held apart by the cross
girder. The moment generated in the restraint is that determined from 2.5% of the moment induced
compression force in the critical flange of the main girder acting at a lever arm corresponding to the
vertical distance between the centroid of the bottom flange (ie the tension flange) and the centroid of the
cross girder. This moment has to be included in the connection design in conjunction with the vertical
shear on the cross girder from applied loading. It also has to be considered in the cross girder design;
however, where this is designed to resist the applied loading as simply supported, the maximum positive
80
Detailed design
bending moment from applied loading will be near mid-span, which is where the restraint induced
moment will be zero, so the restraining actions will not increase the maximum design bending moment
on the cross girder. There is also the opposite case to consider, in which the two critical flanges try to
move towards each other and put the cross girder into compression; however, that case is less severe
than the above case for the connections between the main girder and cross girder and will only govern
the cross girder design if the slenderness ratio for compression buckling about its minor principal axis is
high, generating a low member compression capacity.
3
Note that as described in section 7.3.6 for a pair of main girders restrained by cross girders this detail
provides partial (P) cross section restraint only when there is no slab or plan bracing between the two
beams present.
When the cross girders are more closely spaced than is required to ensure that M* Mbx, then a lesser
restraining force and moment may be designed for in accordance with clause 8.4.2. Further guidance on
the design of restraint is given in section 7.6 below.
7.4
Shear resistance
7.4.1
Shear capacity
The girder web is assumed to be the component that resists the shear force acting on the given member;
in a composite section, the slab is assumed to resist no shear (see clause 6.5). The calculation of the
nominal shear capacity of the web (Vv) is given by clause 5.10 as:

(Equation 7.18)
The nominal shear capacity (Vv) is governed by the shear stress distribution in the web, which can be
either approximately uniform or non-uniform.
For approximately uniform shear stress distribution, Vv = Vu and the value of Vu is dependent on the web
panel depth to thickness ratio (clause 5.10.2). Approximately uniform shear stress distribution occurs in
girders with equal flanges, such as in Universal Beam and welded beam sections.
If the web depth to thickness ratio does not exceed a limiting value (given by clause 5.10.2(2)) then the
nominal shear capacity (Vu) is provided by the gross section area of the web and the value is equal to the
nominal shear yield capacity (Vw). However, if the ratio exceeds the limit then the nominal shear capacity
depends on the nominal shear buckling capacity (Vb).
An unstiffened slender web is unable to develop full shear yield resistance because its capacity is limited
by shear buckling. By introducing transverse stiffeners, the buckling resistance of the web can be
increased. The increase arises firstly from the constraint of the rectangular panel (between stiffeners and
flanges) and secondly because in very slender panels, some of the shear is carried by tension field action.
The nominal shear yield capacity (Vw) is given by clause 5.10.4 and the nominal shear buckling capacity
(Vb) is given by clause 5.10.5.
The same rules for calculating shear resistance apply to webs in both compact and non-compact sections
but the webs in compact sections are likely to be less slender, since the depth in compression is limited.
However, although mid-span sections of composite beams may well be compact, since there is very little
depth of web in compression, the web may be sufficiently slender that the shear capacity must be based
on the shear buckling capacity.
81
7.4.2
Moment/shear interaction
Clause 5.11 gives two methods for taking account of combined moment and shear in a beam. The first is
the proportioning method (clause 5.11.2); this method says that if the bending moment is assumed to be
resisted by the flanges alone, then full shear capacity of the web is available to resist the design shear
force.
The second method, the shear and bending interaction method (clause 5.11.3) says that when the design
bending moment is assumed to be resisted by the whole cross section, then the nominal shear capacity in
the presence of bending moment is modified. The modified nominal shear capacity (Vvm) depends on the
ratio (M*/Ms), where (M*) is the design bending moment and (Ms) is the nominal section moment capacity
(note that this is the only place where the value of (Ms) is required for non-compact sections and for a
composite section built in stages the value for the section at the stage considered should be used). An
illustration of the limiting envelope for interaction of shear and bending is shown in figure 7.10.
Additional guidance on stiffened web panels that are designed to resist a combination of bending
moment, shear, axial and transverse loading is given in appendix B of AS 5100.6 (SA 2004).
Figure 7.10
Moment/shear interaction
Vm/Vv
Shear Force
1.2
1
0.8
0.6
0.4
0.2
0
0
0.2
0.4
0.6
0.8
M*/Ms
1.2
Bending Moment
The moment/shear interaction should be checked for the worst moment and worst shear anywhere within
the web panel length (ie between intermediate web stiffeners), rather than at a single section. This is only
slightly conservative for web panels adjacent to internal supports but seems rather onerous for mid-span
regions of beams of compact section, where the panel could extend between stiffeners at the points of
contraflexure (mid-span moment is then considered in combination with shear at the zero moment
position).
For beams constructed in stages, the moment acting on the section should be taken as the total moment
for sections designed as compact (clause 5.1.6), but for sections designed as non-compact an effective
bending moment must be derived for use in the interaction formulae. This is obtained by multiplying the
extreme fibre total stress by the modulus for that fibre in the section that is appropriate to the stage of
construction being checked (clause 6.3.4). The designer should take care to ensure that the fibre for which
total design stress is used to determine the equivalent bending moment is the same fibre that determines
the bending resistance.
7.5
SLS checks for compact and non-compact beams
Clause 3.3 outlines the different SLS checks that must be considered when designing the different
components in the bridge structure and in this case the main girders are being checked for whether they
are non-composite bare steel or composite, compact or non-compact. This guidance is also applicable to
other structural members, such as beams used in restraints. The SLS checks required are:
82
Detailed design
deflection limits for girders (clause 3.3.1)
vibration of girders (clause 12 of AS 5100.2 (SNZ 2004))
stress limits during staged construction (clause 4.6.3)
crack control of slabs in tension (clause 2.4.4 of NZS 3101 (SNZ 2006)).
The common factor between all the above SLS checks is the applied load acting on the structure. When the
SLS is being considered, most of the applied loads are un-factored compared with the ULS case where load
factors are included. The SLS load combinations are given in section 6.4.7.
Other SLS checks for bridge components such as deck slab, bolts and longitudinal shear connections will
be given under their relevant section.
7.5.1
Deflection limits for girders
The deflection limit for girders under traffic loading for SLS is given in clause 6.11 of AS 5100.2, where it
is stated that the deflection shall not be greater than L/600 of the span or L/300 for the cantilever
projection. In addition to the above requirements, the following criteria should be observed which are also
given in clause 6.11:
Deflections do not infringe on clearance diagrams.
The deflection in the hogging moment region (between the supports and point of contraflexure)
should not exceed L/300 of the span.
No deflection is to occur under permanent loading in the sagging moment region.
The deflection is determined by the use of elastic analysis, in accordance with clause 4.2 and as
summarised in section 6.1, for the appropriate SLS loading combination.
7.5.2
Vibration of girders
Section 3.4.15 of the Bridge manual requires that all highway bridges should be checked for vibration.
Although the Bridge manual presents acceptability criteria for vibrations it does not, however, provide
guidance on how this criteria should be verified. To remedy this situation, some recommendations are
presented below.
7.5.2.1
Vehicular traffic
When considering vibrations occasioned by vehicular traffic, although most international standards
describe the severity of vibrations in terms of acceleration, the Bridge manual specifies that the maximum
vertical velocity during a cycle of vibration should be limited to 0.055m/s when the design load is taken as
two 120kN axles of one HN load element. From vibration theory, the velocity of a structure can be
calculated from the following equation:
Where:
fo
= 4
is the natural frequency of the structure
(Equation 7.19)

is the displacement amplitude, which is defined as:
(Equation 7.20)
Where:
83
is the amplitude of the force
is the stiffness of the structure under consideration
is the damping ratio.
Some typical damping ratios, from BS 5400-2 (BSI 2006), for bridges are presented in table 7.2 below:
Table 7.2
Typical damping ratios for bridges

Logarithmic decrement
Damping ratio
Steel with asphalt or epoxy surfacing
0.03
0.477%
Composite steel-concrete
0.04
0.637%
Prestressed and reinforced concrete
0.05
0.796%
Bridge superstructure
7.5.2.2
Pedestrian and cycle traffic
When pedestrian and cycle bridges are considered, the Bridge manual refers to appendix B of BS 5400-2
and appendix A of BD 37/01 (Highways Agency 2010); the latter is available as a free download. Should
the fundamental frequency of horizontal vibration of the bridge be found to be less than 1.5Hz limit, a
dynamic analysis should be undertaken to evaluate the maximum horizontal acceleration in accordance
with BS EN 1991-2 Eurocode 1: Actions on structures part 2: Traffic loads (CEN 2003). According to the
Bridge manual, suitable limits for lateral acceleration are given in Guidelines for design of footbridges (fib
2005).
For cases when the spans of pedestrian and cycle bridges exceed 30m, aerodynamic effects may be
critical. In these cases, wind vibration effects as detailed in BD 49/01 (Highways Agency 2001) should be
considered.
7.5.3
Stress limits during staged construction
Clause 4.6.3 requires that the SLS stresses at each stage shall be calculated elastically and added together.
In sections 5 and 6 of AS 5100.6, no limits are given for total stresses at SLS but it may be presumed that
the limit is the yield strength for steel elements. For non-compact sections, the ULS limit is always more
onerous. For compact sections, the SLS limit may govern in rare situations.
The design steps for this verification are summarised in figure 7.11.
84
Detailed design
Figure 7.11
Verification of stress limits at SLS during staged construction
Determine the effective section, allowing for holes, slender webs
Modify the shear lag factor for
and shear lag
concrete flanges in tension.
Clause 4.4.1
Calculate and add the stresses in each flange at each stage of
Consider the effects due to vertical
construction
loads, shrinkage, differential
Clause 4.6.3
temperature, imposed displacement

and shear lag.
Clause 4.6.3
Redesign
No
Stresses
<
Yes
Go to figure 7.12
limits?
7.5.4
Crack control of slabs in flexure and tension
The requirements for controlling crack width for the concrete deck slab in tension are given in clause
2.4.4 of NZS 3101. The calculated crack width shall not exceed those specified in section 4.2.1(a) of the
Bridge manual unless, alternatively, the requirements of clause 2.4.4.1(a) of NZS 3101 are satisfied.
Notwithstanding this, experience has shown that due to the steel girders preventing the concrete from
shrinking, large restrained shrinkage cracking occurs. In this case, it is recommended that the minimum
reinforcement requirements of clause 6.1.4 of AS 5100.6 are provided to minimise this effect.
7.6
Design of restraints to beams
The restraints to beams need to be designed for the forces generated by their restraining function in
providing restraint against buckling and any directly applied forces. The forces associated with the
provision of restraint are determined in accordance with section 8 of AS 5100.6 and are discussed below.
Directly applied forces are discussed in section 7.6.7.
The forces in the individual bracing members (at a particular location) due to the restraint forces and
applied forces may be determined from a simple frame analysis. The members should be verified for the
bending and axial forces: requirements for tension and compression members are given in sections 9 and
10 of AS 5100.6 respectively and are not discussed in this guide. Advice on the design of tension and
compression members is available in the Steel designers handbook (Gorenc et al 2005) and Simplified
design of steel members (Bird and Feeney 1999).

It should be noted that the restraint stiffness requirement is not given in AS 5100.6. Commentary clause
C8.4.2 of AS 5100.6 states that research conducted by Mutton and Trahair (1975) found that the stiffness
requirements of restraints are satisfied when the strength requirements are met. Similar results were
found by Medland and Segedin (1979). It is concluded that the strength requirements given below will
achieve adequate stiffness for the types of restraint to the main girders.
85
7.6.1
Restraints against lateral deflection
For restraints that prevent lateral deflection in the main girder, clause 8.4.2 states that the transverse
force acting on that restraint shall be taken as 0.025 times (ie 2.5% of) the maximum force in the critical
flange in the main girder (ie for the most onerous ULS load combination) (see section 6.4.7). Where the
restraints are more closely spaced than is required, then clause 8.4.2 allows the forces on individual
restraints to be reduced. However, it is much easier (and not significantly more onerous) to design all
restraints for the 2.5% of the force in the flange.
7.6.2
Restraints against twist rotation
The requirements for restraint against twist about the longitudinal axis of the main girder are given in
clause 8.4.3. For a restraint to be considered as fully effective, the clause requires it to be able to transfer
2.5% of the maximum force in the critical flange in the main girder. A torsional restraint is actually
required to provide a restraining moment, rather than a force. It is presumed that the restraining moment
is due to a couple of the horizontal forces at each flange, the magnitude of the force being 2.5% of the
force in the critical flange. (Again, this is for the most onerous ULS load combination for the location
considered).
For restraints to be considered as providing partial restraint against twist rotation, they must be able to
provide an elastic restraint against twist rotation without rotational slip. Flexible elements, such as
unstiffened webs, may form part of such a restraint provided they are designed and connected to prevent
rotational slip. No advice is given in clause 8.4.2 about the design force for such restraints but it would be
sensible to design them for the same restraint forces given above.
Restraints that permit rotational slip are deemed to be ineffective in restraining twist rotation of the main
girder.
Guidance on the effects of the stiffness of torsional restraints at supports on the elastic buckling moment
is given in appendix A.5 of AS 5100.6.
For intermediate torsional restraints, as discussed in section 7.3.6, their stiffness is taken into account in
determining the slenderness; the design forces associated with provision of that restraint are given in
PD 6695-2 (BSI 2008) and appendix D of this guide.
7.6.3
Parallel restrained members
Clause 8.4.4 gives the requirements for the design forces in restraints to a series of parallel members.
This is specifically for multi-girder bridges where the restraints are connecting all the main girders. In this
case, each restraint shall be designed to transfer transverse force with the sum of 2.5% of the flange force
from the connected member and 1.25% of the flange force from the adjacent connected member.
However, no more than seven members need to be considered. Note that shallow bridges and/or very
wide (>15m) decks may have a large number of main girders, depending on their requirements, in
comparison with typical bridges with three to four main girders with a deck with less than 15m. See figure
7.12 below for details.
Figure 7.12
Restraint being
considered
Parallel restrained members transverse force transfer
2.5%
1.25%
1.25%
1.25%
1.25%
86
1.25%
1.25%
Detailed design
7.6.4
Restraints against lateral rotation
The requirements for effective restraints against lateral rotation (in the plane of the restrained flange) are
given in clause 8.4.5. Stiffness requirements for such restraints are given in appendix A of AS 5100.6 but
no requirements are given in clause 8.4.5 for the values of design moment at the restraint position. As a
rule, a transverse member providing rotational restraint in plan should have the same stiffness as the
stiffness measured about the strong axis of the flange of the girder to which the rotational restraint in
plan is being provided, and should have a rigid connection to that flange. These conditions will seldom be
provided, which is why the influence of rotational restraint in plan is usually ignored. Where restraint
against plan bending is assumed, the restraining members will need to be designed for bending in plan as
well as for axial force due to the lateral restraint provided at the location.
7.6.5
U-frame restraint and decks not at compression flange level
Clauses 8.4.6 and 8.4.7.2 outline the design procedure to determining the horizontal forces acting in the
restraints, in U-frame restraints and for decks not at the compression flange level. These horizontal forces
are the nominal horizontal force (F*u) and the additional nominal horizontal force (F*c). Both types of
restraints follow the same design steps which are summarised in figure 7.13 below.
Figure 7.13
U-frame and decks not at compression flange level design flow chart
Determine form of restraint and select suitable bracing members
Section 7.3.6
Determine effective length (L ) for lateral-torsional buckling
Determine the maximum
Clause 5.6.3
compressive stress in the flange
Calculate the nominal horizontal force (F* )
(f*f ), equivalent buckling stress in

the flange (f ) and (p)
Clause 8.4.6.3 or 8.4.7.3
Clause 8.4.6.2
ci
Calculate the additional nominal horizontal force (F* )
Rotation in radians of the deck at
Clause 8.4.6.2 or 8.4.7.2
its junction with the main beam
Design axial force, N*=F* + F*
under consideration, under the

loading used in calculating (f*f)
Clause 8.4.6.3
Calculate require strength to resist N* as a tension or compression
member
Section 9 or 10 of AS 5100.6
N*N
Redesign
No
Or
N*N
Yes
Satisfactory
7.6.6
Deck at compression flange level
Clause 8.4.7.1 states that if the concrete deck is at the compression flange level and complies with clause
5.4.2, then it can be considered to provide continuous restraint to the critical flange. Therefore, the
87
transverse force is taken as 2.5% (or 0.025) of the maximum force in the critical flange and shall be
distributed uniformly along the span of the main girder. This in turn will be resisted by the shear
connectors between the steel girder and the concrete deck. The design of the shear connectors is given in
section 7.7, using the calculated transverse force to verify this case.
7.6.7
Directly applied forces
Intermediate restraints may be required to transfer forces due to wind (usually very modest magnitude for
medium span bridges) and possibly due to impact forces due to collision on a superstructure. In the event
of seismic action, they may also be required to transfer forces due to horizontal acceleration. If girders are
curved in plan, they will also be required to transfer radial forces due to the curvature.
Restraints at supports are required to transfer all the effects of horizontal forces to laterally restrained
bearings or shear keys.
The effects of seismic action on restraints are not covered in AS 5100.6. However, consideration of
seismic action is required in New Zealand and for restraints the check is straightforward. In this case, the
horizontal force due to seismic action is determined from global analysis (using computer modelling
described in section 6.3 for the seismic load combination (section 6.5) acting on the restraint. This in turn
is compared with either the transverse force or the nominal horizontal force, as described above, and the
maximum calculated force is used to design the restraint.
7.6.8
Restraints at supports
Bracing or pier diaphragms at supports provide torsional restraint to the main girders and a load path for
transferring lateral forces to restraint bearings and/or shear keys. The same 2.5% transverse force and
directly applied forces should be considered at the supports when designing restraints at supports.
7.7
Longitudinal shear connection
To provide the necessary shear transfer between the steel girder and the concrete slab that is required for
composite action, shear connectors are required on the top flange. The shear flow varies along the length
of the beam, being highest near the supports, and it is customary to vary the number and spacing of
connectors to provide just sufficient shear resistance for economy. The most commonly used form of
connector is the headed stud, though channel connectors and high strength structural bolts may be used
(clause 6.6.4.1).
The SLS, fatigue and ULS design procedures for longitudinal shear connection are set out in figures 7.14 to
7.16.
88
Detailed design
Figure 7.14
Longitudinal shear connection SLS design flow chart
Calculate the design longitudinal shear force per unit length (v* ) at
L
supports, mid-span and points of contraflexure
global analysis. Un-cracked

section properties.
Clause 4.4.3
Select connector size and determine nominal strength (v )

Ls
No
SLS design shear force (V*) from
Characteristic shear capacity of the

connector (f ks)
Clause 6.6.4
v* v
L
Ls
Yes
Select connector spacing to provide required design resistance

Check fatigue
Figure 7.15
Figure 7.15
Longitudinal shear connection fatigue design flow chart

Calculate stress range in weld to connector
Section 7.7.6
Calculate longitudinal shear force under

fatigue loading. Un-cracked section
properties
Determine the fatigue strength (f ) for the given weld detail

f
Clause 13.6.2 or 13.6.3 for headed studs/section 7.7.6
Clause 4.8 of AS 5100.6 and clause 6.9

of ASthe
5100.2/section
7.7.6stress
Determine
number of fatigue
cycle
Clause 6.9 of AS 5100.2/section 7.7.6
Redesign
Figure 7.14
Stress range
No
fatigue
strength (f )
f
Yes
Check ULS
Figure 7.16
89
Figure 7.16
Longitudinal shear connection ULS design flow chart (or design of transverse reinforcement at shear
studs)
Calculate the design longitudinal shear force per unit length (v* ) at
Lp
supports, mid-span and points of contraflexure and at positions where

the provision of connectors changes
ULS design shear force (V*) from global

analysis. Un-cracked section properties.
Clause 4.4.3

Calculate the design capacity for longitudinal shear.

Modify the shear capacity for coexistent transverse bending
Determine minimum transverse reinforcement and its curtailment
Clauses 6.6.5.4 or 6.6.5.5 and 6.6.5.6/section 7.7.6
Redesign
Figure 7.14
No
Are all ULS

checks
Yes
Satisfactory
satisfied?
7.7.1
Longitudinal shear
The longitudinal shear is the means by which axial force is transferred from the girder into the slab. The
design longitudinal shear force per unit length (v* )L is given in clause 4.8, based on simple elastic theory
of bending, by the following expression.
Where:
(Equation 7.21)
V* = design shear force at the cross section.

Clause 4.8 does not distinguish between design shear forces at ULS and SLS; the expression applies to
both situations. The expression is presumed to apply at ULS even when the cross section is compact (and
thus the plastic bending capacity is mobilised).
Where the beam cross section varies significantly (eg for a tapered section or for different girder
dimensions at the hogging and sagging moment region) the variation of section properties must be taken
into account (use the properties at the cross section being considered).
Clause 4.8 states that the design longitudinal shear force is to be calculated using the un-cracked section
properties in sagging moment region with the short-term concrete modular ratio, while for the hogging
moment region it allows either the use of the area of the embedded steel and the un-cracked concrete
area or the area of the embedded steel on its own. This gives a conservative value without the need for
more complex calculation, even when the plastic resistance of the cracked section is relied upon. Short
term un-cracked properties may be used for this purpose. This force is calculated for both the SLS and ULS
cases using the relevant design shear force (V*) for each case. This is based on the load combination
90
Detailed design
summarised in section 6.4.7, which not only includes the traffic loading but other loading such as
temperature effects and shrinkage, especially at the hogging moment region.
In the positive moment regions, the design (v* ) varies from a maximum at the point of contraflexure to
L
zero at the point of maximum moment. This can be used to reduce the shear stud numbers (by increasing
the spacing) accordingly.
7.7.2
General requirements
Clause 6.6.1 requires the shear connection to be designed to satisfy SLS requirements and for fatigue
resistance (section 7.7.4). No checks are required at ULS, except when the connectors are also subject to
tensile forces (note that the reference to clause 6.6.3.3 in clause 6.6.1 is erroneous). The clause also
requires the transverse reinforcement to be designed to resist the longitudinal shear force but for this
check, ULS values of design longitudinal shear force must be used.
The values of characteristic shear resistance given by clause 6.6.4 are only valid when certain dimension
limitations on spacing and height of the shear connectors are met. These limitations are set out in
clause 6.6.2.
7.7.3
Design for shear at SLS
The basic requirement for shear connection is expressed in clause 6.6.3.2 as:
(Equation 7.22)
Where:
v* = design shear force per unit length,

L
= permissible shear force per unit length,
ls
= capacity reduction factor (given by table 3.3 of AS 5100.6 as 1.0).
Over lengths where the shear connector spacing is constant, clause 6.6.3.2 allows the SLS value of shear
per unit length at any particular location to exceed the capacity per unit length (vLs) by up to 10%,
provided that the total shear over the length where the spacing is constant does not exceed the total
design shear resistance over that length.
For negative (hogging) moment regions, clause 6.6.3 imposes a further requirement that the adequacy to
resist a total horizontal shear force (F* h) should also be verified. This force will be resisted by the shear
connectors between the position of maximum negative bending moment and the adjacent position of zero
moment (the point of contraflexure), at the SLS case. Therefore, in the hogging moment region the more
onerous of the longitudinal shear force (v* ) (as it and, potentially, the capacity vary along this length) and
L
the total horizontal shear (F* ) shall be used when designing for the SLS case. The total horizontal shear
h
(F* ) expression is given in clause 6.6.3.3 which is outlined below:

h
= 0.55
(Equation 7.23)
Where:
Ars = area of slab reinforcement within the effective width of the slab
fsy = yield stress of tensile reinforcement.
91
7.7.4
Design for shear at ULS
In certain cases, shear connectors may be subject to tensile force as well as to longitudinal shear. This
situation arises when forces tend to separate the slab from a girder; or when there are transverse
moments on a group of connectors resulting from the restraint to transverse bending of the slab,
particularly in the region of diaphragms or transverse cross-bracing as stated in clause 6.6.3.4. In the
latter case, additional ties, suitably anchored, are required to resist these forces.
For such situations, the requirement at ULS is expressed as a limit to the reduced shear capacity:
Where:
< 3
= capacity reduction factor (given by table 3.2 of AS 5100.6)
= number of shear connectors per unit length
(Equation 7.24)
fks = characteristic shear capacity of the connector specified in clause 6.6.4

N*u = design axial tension on the shear stud at the strength limit state.
The requirement for fatigue resistance is also modified in these situations. See clause 6.6.3.4(b)(ii).
7.7.5
Positioning of shear connectors
For the main girders, if shear studs are used, they are typically set in groups of 2, 3 or 4 across the width
of the flange; the spacing, and sometimes the number across the width, varies in a series of steps along
the beam, with wider spacing and/or fewer studs per row in regions of low shear. For cross girders in
ladder decks, the flange is often too narrow for more than two studs across the width.
The spacing of studs (for main girders and cross girders) needs to be coordinated with the spacing of
transverse reinforcement, to avoid potential clashes in positions. Detailed requirements for the placement
of the shear connectors are given in clause 6.6.2.
7.7.6
Fatigue design of shear connectors
Section 13 of AS 5100.6 covers the fatigue design requirements for all welded components in the bridge
structure, which is discussed in section 7.11. However, the fatigue design of shear connectors is given
below.
It should be noted that other than the specific tension case outlined in section 7.7.4 above, the design of
shear connectors will be undertaken for the SLS case. However, as seen in figure 7.16, if the shear
connector did not satisfy the fatigue requirements, then the connector will have to be redesigned
accordingly. This can either be done by changing the stud spacing first and if needed increasing the
number of shear connectors or using bigger headed studs or bolts.
7.7.6.1
Fatigue loading and number of fatigue stress cycles
Clause 6.9 of AS 5100.2 gives the requirements for determining the fatigue loading on shear connectors
and the number of fatigue stress cycles. This data is required for determining the stress range in the weld
to the shear connector. Further information on the use of the fatigue loading is given in section 6.4.2.
7.7.6.2
Calculation of the stress range (f*)
The stress range (f*) in the weld to the shear connector is calculated as follows:
92
Detailed design
Calculate the longitudinal shear force per unit length (v* ) (clause 4.8), using the fatigue loading
L
(clause 6.9 of AS 5100.2 as modified by Clifton (2007)), at supports, mid-span and points of
contraflexure. Further guidance on the fatigue loading is given in section 7.11.
2
Take the longitudinal shear force as per metre length for the considered region and divide it by the
number of shear connectors in a row (this can be one for a channel connector to up to four for headed
studs or bolts) and the cross sectional area of the shear connector in contact with the girder. The
result is the design stress range, in this case of shear stresses (f* ), of the weld to the shear
s
connectors, given as MPa (N/mm2). This expression is summarised below:
=
Where:
(Equation 7.25)
v* is the longitudinal shear force per unit length

L
is the number of shear connectors
Asc is the cross sectional area of the shear connector.

7.7.6.3
1
Determine the fatigue strength (ff) for the given weld detail
Determine the weld detail for the specified shear connector. This is taken as detail category 80 for
headed studs, detail category 80 for channel connectors (for t 12mm) and detail category 100 for
bolts.
Determine the number of fatigue stress cycle (nsc), see section 7.11 for further details.
Determine the fatigue strength (ff) for the given weld detail and number of fatigue stress cycle, which
is plotted on the S-N curve for shear stress in clause 13.6.2 or use the expression given in clause
13.6.3 for headed studs.
If the stress range of the weld to the shear connector is less than or equal to the fatigue strength (ff),
then fatigue is satisfied. If not then, the shear connector will need to be redesigned, ie:
7.7.7

(Equation 7.26)
Design of transverse reinforcement
The transverse reinforcement in the slab is required to transfer the longitudinal shear forces from the
zone around the connectors to the rest of the slab. The requirements are expressed in clause 6.6.5. The
longitudinal shear force is calculated in a similar manner to that for the shear connectors but the forces
are evaluated on possible planes of failure, called shear planes.
An example of typical shear planes is illustrated in figure 7.17. For a shear plane around the studs, the
design longitudinal shear force at ULS (note the difference that the capacity of the connectors is normally
verified only at SLS) but for shear planes either side of the beam, the shear force additionally depends on
the relative magnitude of the effective width of slab on that side. For a ladder deck bridge, most of the
slab and thus the greater part of the shear force, is on one side.
93
Figure 7.17
Typical shear planes
Att
Abt
Once the design longitudinal shear force is calculated for the particular shear plane, the design capacity of
the transverse reinforcement must satisfy the criteria in clause 6.6.5.2 and, for certain shear planes,
clause 6.6.5.3 must be satisfied. In both cases, guidance on the use of haunches and their design is given
in these clauses. Note that, for cross girders in ladder decks, transverse refers to the cross girder and
thus the reinforcement is the longitudinal reinforcement in the slab; in hogging moment regions that
reinforcement will be under significant tensile stress. Transverse reinforcement is required to provide the
tensile resistance to global bending, local resistance to longitudinal bending and the transfer of
longitudinal shear from the cross girder. Rules for interaction are, however, not given in AS 5100.6 but it
would be pragmatic to provide an area of reinforcement that is at least the sum of the areas needed to
resist each of the effects separately.
Finally, clauses 6.6.5.4, 6.6.5.5 and 6.6.5.6 conclude the transverse reinforcement design by providing
limiting values of minimum transverse reinforcement and its curtailment. This minimum transverse
reinforcement shall also not be less that the value given in clause 9.3.9.4.15 of NZS 3101. This provision
relates to the design of the deck slab, which is discussed in section 7.10 below.
See section 8.2.3 for guidance on detailing the deck slab in the region of the shear connectors.
7.8
Connections
Connections between steelwork components are made by welding or bolting (rivets are rarely used
nowadays). Generally, welding is used for connections made in the fabrication shop (butt welds in flange
plates, web/flange welds, connection at stiffeners etc) and bolting is used for connections on site (using
cover plates, gusset plates, cleats etc). Welding is rarely used on site because of the extra control
processes that have to be made available, the need for weather protection and the additional time
involved, although for large projects it might prove beneficial.
The quantity of weld metal is small in comparison to the mass of the structural components but the cost
of welding is comparable with that of the material. The splice plates and bolts are also small in mass
relative to the structural members but the process of connecting them is a labour-intensive process that
is a major part of the cost of erection. Economy is therefore achieved through simplicity, minimising the
number of components and making the connection process as straightforward as possible. The designer
must keep in mind the following considerations:
Design for strength:

94
Detailed design
do not overdesign the connections
arrange the connections for direct force-transfer, with minimal eccentricities.
Design for fatigue resistance:
avoid poor fatigue details in highly stressed locations
avoid creating notches and other stress-raising details.
Design for serviceability:
avoid details that might cause local yielding under working load
avoid details that would cause difficulty in of application of protection (and other) coatings
avoid features that would cause collection of water and debris.
7.8.1
Requirements for the design of connections
The requirements for the design of connections are given in section 12 of AS 5100.6. The general
requirement stated in clause 12.1 is that connections shall be capable of transmitting the calculated
design action effects. It may be noted that this means that a connection does not necessarily need to be as
strong as the components that it connects. However, clause 12.3.1 sets out minimum values of design
actions for connections, many of which are related to the resistance of the connected member (typically
50% or 75% of the nominal member capacity).
Detailed rules for the design of bolted and welded connections are given in clauses 12.4 to 12.6. Guidance
on bolting and welded design procedures is available in Gorenc et al (2005); Clifton (1994); or SCNZ
(2007); the first of these references refers to AS 5100, the second and third references refer to NZS
3404.1 (2009) although the design procedures are similar in both standards. The following sections
summarise the key aspects of the bolting and welding design procedures.
7.8.2
Bolting
Most bolted connections in bridges transfer forces between parts by means of bolt shear. Tension
connectors are rarely used to transmit loads between primary components.
Clause 12.5.1 defines four types of bolting category as shown in table 7.3.
Table 7.3
Bolting category
Bolting category
Bolt grade
Method of tensioning
Means of load transfer
4.6/S
4.6
Snug-tight
Bearing/shear
8.8/S
8.8
Snug-tight
Bearing/shear
8.8/TB
8.8
Full tensioning
Bearing/shear
8.8/TF
8.8
Full tensioning
Friction at SLS
Bearing/shear at ULS
Clause 12.3.2 states that all connections and splices in the main girder shall use bolting category 8.8/TF
(also known as fully tensioned property class 8.8 bolt to AS/NZS 1252 (SNZ 1996)). Although this would
seem to permit the use of other bolting categories for connection of bracing members it is preferable to
use only bolting category 8.8/TF bolts for all connections of structural members in bridges.
Bolts should be positioned in accordance with the limits given in clause 12.5.2.
95
Where weathering steel is used, it is common practice to design the connection for M24 bolts but to set
out the bolt positions according to minimum spacing and edge distance values for 1 inch size bolts
because the bolts are likely to be supplied from the USA in that size. Additional guidance on the use of
weathering steel is found in El Sarraf and Clifton (2005).
Each bolted connection has many bolts. A key consideration in design is therefore the sharing of load
between the bolts in the group. In some cases a quasi-elastic linear distribution of forces between bolts is
used (ie load per bolt is proportional to its distance from a centre of rotation) and in some cases a plastic
distribution can be used (ie the full resistance of each bolt is used).
The design resistance of a bolted connection is based on the resistances for individual bolts. Bolt capacity
at ULS must satisfy the provisions in clause 12.5.3 and slip resistance at SLS must satisfy clause 12.5.4.
Bolting category 8.8/TF bolts are usually used in normal sized holes but the designer may wish to allow
the use of oversize holes in some locations, for constructional reasons. If oversize holes are to be
accepted, this needs to be recognised in design, since the design resistances are reduced and the
minimum hole spacing is increased (because it is based on hole diameter).
Additional guidance on the design of a bolted splice is given in section 7.8.4 below. Note that property class
10.9 bolts are also available and can be considered in connections that require additional bolt capacity where
space is limited for instance. However, this type of bolt is susceptible to hydrogen pickup, possibility leading
to delayed brittle fracture. Specialist advice should be sought before specifying these bolts.
7.8.3
Welding
Welded connections are made using either fillet welds or butt welds and clause 12.6.1 states that all
welding shall be carried out in accordance with AS/NZS 1554 (SNZ 2004), specifically part 5.
The strength of a fillet weld depends on its throat thickness (see figure 12.6.2.4(a) of AS 5100.6) and for
some welding processes this throat thickness can be increased by extending the fusion zone: this is then
referred to as a deep penetration fillet weld (see figure 12.6.2.4(b) of AS 5100.6). Butt welds are
categorised as either complete penetration butt welds or incomplete penetration butt welds, depending on
whether the fusion zone extends through the complete or only partial depth of the joint. Clause 12.6.3.1
only permits incomplete penetration welds for longitudinal welds such as the web/flange welds. All welds
shall be designed to clause 12.6.7 and shall comply with the requirements of AS/NZS 3679.2 (SNZ 2010b),
especially table 3 of that standard.
The inspection and testing of welds is an important aspect of quality control in fabrication; see NZS
3404.1 (SNZ 2009) for further advice.
7.8.4
Splices in main girders
For all but short single spans, each main girder is fabricated in a number of pieces and joined together on
site, either prior to or during erection. The lengths of the pieces are usually chosen to suit economical
fabrication and transport restrictions, and splice positions are usually arranged to be away from positions
of maximum moment. As noted above, the splice may then be designed to transmit the most onerous
design force and moment at that position, which is likely to be significantly less than the full design
resistance of the girder. Splices may either be bolted or welded.
7.8.4.1
Bolted splices
For bolted splices, cover plates are normally provided on both faces of each flange and web. The number
of bolts required is determined on the basis of no slip at SLS and bearing/shear resistance at ULS. When a
96
Detailed design
connection is assumed to act in bearing/shear at ULS, the requirement for no slip at SLS usually governs
(the chief exception being for thin material when large bolts are used).
In a bolted splice, a key design task is to determine the distribution of forces between the individual bolts.
For each flange, the number of bolts in the connection can be determined on the basis of the flange force,
calculated on the basis of an elastic stress distribution in the girder section. Therefore, at ULS, the
resistance of the bolt group is the sum of the resistances of all the fasteners.
For the web, the force on the web plate connection, calculated from an elastic stress distribution, is a
combination of moment, axial force (the centroidal axis of the section is not usually at mid-depth in the
web) and shear. The shear is transferred across the connection; this imposes both a shear force and an
additional moment (shear force times eccentricity) on the bolt group. The force in each bolt is the vector
sum of the following:
the vertical force due to sharing the shear force equally between all the bolts
the horizontal force due to axial force on the web if required (again shared equally)
the force due to moment on the web (each bolt force is directly proportional to the distance from the
centre of the bolt group and acts tangentially to that radius).
The force on the outermost bolt determines the design of the group. In this case, at ULS the bearing
resistance may govern, rather than slip resistance, because webs and their cover plates are usually thin.
In the flange and web cover plates, stresses should be checked where the plate is in tension or where the
holes are oversize or slotted, allowing for holes for fasteners in determining net sections (see clauses
12.3.6 and 12.4).
On the upper top flange cover plate it may be necessary to provide shear studs, to comply with maximum
longitudinal spacing limitations for shear studs. Only a single row of studs should be provided, if possible,
to avoid complications in tightening the bolts.
For further advice on the design of bolted connections, see Gorenc et al (2005), Clifton (1994) or SCNZ
(2007). A typical bolted splice is shown in figure 7.18.
Figure 7.18
7.8.5
Typical bolted splice
Welded splices
Welded splices in girders usually involve full penetration butt welds in webs and flanges. As noted above,
full penetration welds should always be used: this makes the splice capable of the same bending
resistance as the girder section.
97
The butt weld in the web is often staggered relative to that in the bottom flange, to assist in locating the
web during erection (the web can sit on the projecting length of flange). A semi-circular cope hole is
usually provided in the web above the flange weld, to facilitate the butt welding of the flange. Where the
splice is not staggered, the cope hole is usually filled after the splice has been welded, to avoid the stress
concentration (around the open hole) at the end of the butt weld in the web.
Details are shown in figure 7.19. Where the cope hole does not have the termination of a butt weld it may
be left open.
Figure 7.19
Arrangement at welded splice
7.8.6
Cross girder end connection
7.8.6.1
Design basis of intermediate cross girders
During construction, there will be very little end moment on the cross girder due to gravity loads (ie there
is very little end fixity from the main girder) and the single lap connection is assumed to transmit the
vertical shear and a sagging moment equal to the shear multiplied by the distance from the web to the
centroid of the bolt group.
Once the slab has been cast, the connection will act compositely with the deck slab. The connection will
need to transfer the vertical shear (acting on the line of the web) and moments about the main girder axes
due to U-frame action as a result of differential loading on adjacent cross girders and the restraint
provided to the compression flanges of the main girders in the hogging moment regions.
Hogging moments from the deck slab cantilever reduce the sagging moment to be transmitted at the
centroid of the bolt group. Such moments disperse from moment carried by the slab alone to moment
carried by the composite section over the length of the first few stud connectors and can be assumed to
act on the composite section at the position of the bolt group. However, unless the governing design case
for the bolt group is with net hogging moment, the cantilever slab moments can be neglected.
The bolted connections provide restraint to the main girder against lateral-torsional buckling in regions
adjacent to the intermediate support and can be designed for bolting categories 8.8/TB using the ULS
restraining forces. This is also applicable to the mid-span regions.

The forces on the bolts in the composite condition are determined by considering a Tee section
comprising a width of slab plus the web of the cross girder. There are no effective width rules for this
98
Detailed design
design situation but it is suggested that a width of slab equal to the width of the main girder flange will
suffice for the connection design. The horizontal force and moment on the web should be determined
from the stress distribution in this Tee section and the bolt group then designed for the combination of
shear, horizontal force and moment. The force on the bolt in the bottom row will govern.
See section 9.2 for further details.
7.8.6.2
Design basis of pier diaphragms
In the case of pier diaphragms, a similar design basis to that for the intermediate cross girders is adopted,
though it could be argued that the crosshead connection may be designed to slip into bearing under
bending and shear at ULS as the flexibility that this creates has little effect on the lateral-torsional buckling
of the main girder (a more flexible end support would need to be assumed in evaluating the buckling
resistance but the effect is small).
In this splice configuration the bolt group to be considered is the one on the crosshead side of the splice;
it carries greater moment due to its eccentricity from the main girder.
The use of a stub flange, with bolts in the cover plates at maximum distance from the slab, is more
effective in transmitting moment; the stub flange then has to be designed to transfer the forces into the
stiffener. See section 9.2.2 for further details.
7.9
Stiffeners
Web stiffeners are required to improve the shear resistance of the web, for the attachment of transverse
bracing or cross girders and as support over bearings. In some cases, they can also be used to improve
the compression resistance of a slender web, as well as, the effectiveness of the cross section. Three
different types of web stiffeners are covered in AS 5100.6, they are:
load-bearing stiffeners
intermediate transverse web stiffeners
longitudinal web stiffeners
The main considerations in the design of web stiffeners, which must be satisfied, are discussed below.
Further detailed guidance on the design of stiffeners is available in Gorenc et al (2005) and Clifton (1994).
7.9.1
Load-bearing stiffeners
Clause 5.13 gives the requirements for the design of load-bearing web stiffeners.
Load-bearing stiffeners are used to strengthen the web to carry concentrated forces such as the reactions
at supports. Bearing stiffeners are usually flat plates provided on both faces of a web, and are usually
positioned symmetrically about the web, so that the centroid of the effective section is on the line of the
web. In some cases, this would give resistance to bending about a transverse axis while allowing a single
lapped connection to transverse bracing. See figure 7.20 for typical details. The ends of the stiffeners are
normally fitted to bear against the flange where the concentrated load is introduced and fillet welded to
the web and flanges.

Clause 5.13.4 requires that the stiffeners shall be provided with sufficient welds [or bolts] to transmit the
entire bearing force or design reaction (R*) to the web. This requirement does not preclude the
transmission of (part or all of) the reaction to the end of the stiffener in bearing and it is common to
design that connection for direct bearing at ULS, when the stiffener has been fitted. However, it should be
noted that for fatigue design the variation in reaction due to the passage of the fatigue load should be
99
assumed to be transmitted through the welds, not in bearing. Note that clause 5.13.4, despite its heading,
does not make any statement about the fitting (as opposed to attachment) of the stiffeners.
Figure 7.20
Typical bearing stiffeners
Load bearing
Main girder
stiffeners
Elevation
The stiffener outstands from the face of the web (bes) need to comply with the geometric limit given in
clause 5.13.3.
15

250
(Equation 7.27)
Where:
bes
is the stiffener outstand from the face of the web
ts
is the flat stiffener thickness
fys
is the design yield stress of the flat stiffener without the outer edge continuously stiffened (eg the
stiffener is from a flat bar or plate and not an angle section).
Load bearing stiffeners must be designed at ULS for yield capacity (clause 5.13.1) and buckling capacity
(clause 5.13.2); the requirements are expressed respectively as:
yield capacity:
(Equation 7.28)
buckling capacity:
Where:
(Equation 7.29)
R* is the design bearing force or design reaction, including the effects of any shear forces applied
directly to the stiffener.
is the capacity reduction factor (given in table 3.2 of AS 5100.6)
Rsy is the nominal yield capacity of the stiffened web, given in equation 5.13.1(2) of AS 5100.6.
Rsb is the nominal buckling capacity of the stiffened web and stiffener.
The calculation of the nominal buckling capacity of the stiffened web and stiffener (Rsb), given in clause
5.13.2 requires further guidance. The recommended procedure is:
100
Detailed design
Determine the effective section (As) of the compression member is taken as the stiffener area plus an
effective length of the web, taken as the lesser of:
17.5
(Equation 7.30)
250
Where:
tw
is the web thickness.
is the web panel width or spacing to the next web stiffener, if present.
Determine the effective length (Le) of the compression member, is given in equation 5.13.2(2) and (3),
see below:
a
where the flanges are restrained by other structural elements against rotation in the plane of the
stiffener, or
= 0.71
(Equation 7.31)
if either of the flanges is not so restrained
= 1
(Equation 7.32)
Where:
= depth between flanges.
Calculate the compression member slenderness reduction factor () given in clause 10.3.3, taking
c
as 0.5 and kf as 1.0. Note that the radius of gyration (rs) is taken as:
Where:
Is
4
(Equation 7.33)
= the web stiffener second moment of area about the axis parallel to the web.
Calculate the nominal buckling capacity of the stiffened web and stiffener (Rsb) for the parameters
given in clause 5.13.2:
=
(Equation 7.34)
The requirement in clause 5.13.5 for when load-bearing stiffeners are the sole means of providing
torsional end restraint at the support(s) only applies in circumstances such as half-through bridges. For
the forms of bridge described in this guide, bracing would normally be provided at supports even during
the bare steel construction stage.
7.9.2
Intermediate transverse web stiffeners
Clause 5.14 gives the requirements for the design of intermediate transverse web stiffeners, which are
used to increase the buckling resistance of the web and for the attachment of bracing between beams.
Intermediate transverse web stiffeners are usually flat plates fillet-welded to one face of the web and to
one or both flanges. For deep slender webs angle stiffeners, with one leg outstanding, are sometimes
used, for greater stiffness.
Generally, on the outermost beams, the stiffeners should be on the hidden face, rather than the exposed
face, for better appearance; any stiffeners to which bracing are attached would be on the inner face in any
case. To restrain the web against buckling, the stiffener does not need to be connected to either flange;
101
although it is usual to connect it to the top flange (it avoids potential fatigue at the flange/web weld due
to transverse flexure of the slab). At the bottom, the stiffener can be stopped short (subject to a maximum
clearance of four times the web thickness, according to clause 5.14.1); this simplifies fabrication and is
thus more economic.
Where the stiffener acts as a connection for transverse bracing, it should be connected to both flanges.
Practical experience indicates that failure to provide such attachment may lead to fatigue cracking in the
web at the point of curtailment of the stiffener.
Where bracing is attached to a web stiffener, the stiffener may need to be shaped to provide sufficient lap
to connect the bracing members, see section 9.2 for further details.
The fillet welding of the end of a stiffener to a flange does not introduce a lower class of fatigue detail
than is likely to be present already, provided that the toe of the weld is at least 10mm from the edge of
the flange. Web stiffeners should be proportioned such that they are narrower than the flange outstand;
shaped stiffeners for bracing may need to be notched at the end (see figure 7.21) to ensure that the welds
are not too close to the edge of the web.
Figure 7.21
Example of an intermediate transverse web stiffener with a notched end
Notch
Intermediate
transverse
stiffeners
Main girder
Elevation
The outstands of intermediate transverse web stiffeners need to comply with the same geometric limit as
load-bearing stiffeners (see clause 5.14.2) and to have a minimum area (clause 5.14.4), given by:
0.5(1 )
Where:
is the area of stiffener,
is the factor for transverse stiffener arrangement, taken as:
= 1.0 for a pair of stiffeners

= 1.8 for a single angle stiffener
= 2.4 for a single plate stiffener
2
1+
2
(Equation 7.35)
Aw is the area of web
v is the shear buckling coefficient determined in accordance with clause 5.10.5.2

102
Detailed design
Stiffeners are also required to have a minimum stiffness, given by clause 5.14.5 as:
0.751 3
1.513
1
> 2
(Equation 7.36)
(Equation 7.37)
At ULS, stiffeners not attached to bracing are required only to have a buckling capacity greater or equal to
that calculated to clause 5.14.4 as:
(+ )
(Equation 7.38)
Where
is the capacity reduction factor (see table 3.2 of AS 5100.6.
Rsb is the nominal buckling capacity of the web and intermediate stiffener determined in accordance with
clause 5.13.2.
Vb
is the nominal shear buckling capacity specified in clause 5.10.5.2 for a stiffened web using
and = 1.0.
1
= 1.0
Additionally, clause 5.14.8 states that the web connections of intermediate transverse stiffeners not
subject to external loading shall be designed to resist a minimum design shear force per unit length
(kN/mm), of not less than:
0.00082
(Equation 7.39)
Stiffeners subject to external loads or moments are also required to be designed for Increase in strength
(clause 5.14.7.2) and increase in stiffness, given by clause 5.14.7.1 as a minimum value of the second
moment of area (Is):
2+
3
= 1
Where:
(Equation 7.40)
F*p is the transfer design forces

e
is the distance between the end plate and load-bearing stiffener.
It should be noted that due to the nature of loading on intermediate transverse stiffeners, the yield
capacity checks are not required.
7.9.3
Longitudinal stiffeners
Clause 5.15 outlines the design requirements for the design of longitudinal stiffeners. These stiffeners are
sometimes used in very deep girders, as they have a significant effect on the compressive buckling of a
slender web.
For best results, longitudinal stiffeners should be placed at a distance of about 0.2d from the compression
flange; clause 5.15.2 refers only to stiffeners at this location and at the neutral axis. Clause 5.15.1
requires these stiffeners to be continuous or extend between and be attached to transverse stiffeners.
Longitudinal stiffeners at 0.2d are required to satisfy the following stiffness requirement only:
minimum stiffness (clause 5.15.2)
1 +
421 +
103
(Equation 7.41)
Where:
As is the area of the stiffener

d2 is twice the clear distance from the neutral axis to the compression flange area of the stiffener.
Clause 5.15.2 states that, when a second longitudinal stiffener is required at the neutral axis of the
section, it shall have a second moment of area (Is) about the face of the web such that:
3
(Equation 7.42)
7.10 Deck slab design
The design of the reinforced concrete deck slab is carried out in accordance with NZS 3101, rather than
with AS 5100.5 (SA 2004), as discussed in sections 5.3 and 6.3. In this section, references to specific
clauses are to NZS 3101, unless noted otherwise.
Clause 12.8 sets out two acceptable methods of design for reinforced concrete bridge decks:
empirical design, based on membrane action
elastic plate bending analysis.
Due to the stringent acceptance criteria outlined in clause 12.8.2.2, the use of empirical design is not
commonly used. Therefore, the second method is normally adopted and is the only method discussed
here.
As noted in chapter 6, the slab participates with the main girders and (in ladder decks) with cross girders and
the effects due to this composite action are revealed by the global analysis. A separate local analysis is often
used to determine the local effects in the slab due to the individual wheel loads. The slab must then be
verified for the different loading combinations (see section 6.4). Although the slab is predominantly one-way
spanning for global effects, it is two-way spanning for local effects and the rules in section 12 of NZS 3101
apply (with appropriate references to other clauses for detailing requirements).
7.10.1 Slab reinforcement

It is usual to provide two orthogonal layers of reinforcement on each face of the deck slab over the whole
area of the deck. For a slab cast wholly in situ, one layer is usually laid on the bottom; for a deck cast with
partial depth precast units, some of the bars are within the units and the arrangement is shown in figure
7.22. Further information is given in section 9.4.
Figure 7.22
Partial depth precast units reinforcement details
In a multi-girder deck, the greatest bending moments due to the traffic loading are transverse; both
sagging and hogging moments are generated (sagging is usually greater). Global bending due to the
104
Detailed design
differential deflection of the main girders also contributes to transverse bending in the slab.
Consequently, the transverse reinforcement is considered as the principal reinforcement and is usually
placed as the outer layer.
In a ladder deck bridge, the greatest bending moments in the slab due to traffic loading occur in the
longitudinal direction: both hogging and sagging moments are generated. Transverse moments (due to
the local effect of wheel loads) are smaller and mainly sagging in nature the exception is over the length
of the cantilever and immediately inboard of the main girders. Consequently, the longitudinal
reinforcement is considered as the principal reinforcement and usually placed as the outer layer. However,
the reduced lever arm for transverse bending resistance is a disadvantage at the root of the cantilever but
the penalty is normally accepted.
7.10.2 Design for flexure

The basic requirement for design for flexure in slabs is given in clause 7.4.1:
(Equation 7.43)
Where:
M* is the design bending moment

Mn is the nominal flexural strength of that section of the slab
is the capacity factor, given by clause 2.3.2.2 of NZS 3101.
It should be noted that the design for flexure for two-way slab, is according to the requirements outlined
in section 9 of NZS 3101 and clause 12.5.1. The main difference in the two outlined design methods is in
the shear design of the slab, which is given in section 7.10.3 below.
Using one of the analysis models described in section 6.3, such as the commonly used 2D grillage, the
slab and reinforcement are designed as a beam as well. In this case, the slab, or the idealised beam, can
be designed as a one-way slab in accordance with the provisions of section 9 of NZS 3101. To take
advantage of the larger lever arm, the layer of reinforcement that resists the largest local moment is
placed as the outer layer, while the inner layer then resists the lesser orthogonal moment. The positioning
of the principal reinforcement is dependent on the type of bridge being used, which is clarified below.
In either case, the following design requirements must be satisfied when designing for flexure:
Clauses 9.3.6.2 and 9.3.8.1 require that all flexural tension reinforcement shall be well distributed
across the zone of maximum tension and there is sufficient reinforcement to meet crack control
requirements. The amount and distribution of longitudinal reinforcement shall be such that, at every
section, the distance of the extreme compression fibre to the neutral axis is less than 0.75 of the
corresponding distance for balanced strain conditions (c ). Designers should also note the provisions
b
of clause 6.1.4 of AS 5100.6 on the minimum longitudinal reinforcement (this provides the minimum
requirements in the sagging moment regions but it is common practice to provide for this minimum
requirement in the hogging moment region also).
Clause 9.3.8.2.1, outlines the minimum reinforcement area (As) needed in the idealised beam (ie slab),
which shall be greater than that given in the expression below:
> 4 (Equation 7.44)
But equal or greater than:
105
1.4
(Equation 7.45)
Clause 9.3.8.3 outlines the reinforcement spacing in the slab. For the principal reinforcement (as
outlined below for the different bridge types) the spacing is given as not exceeding the lesser of twice
the slab thickness or 300mm. For reinforcement perpendicular to the principal reinforcement the
spacing shall not exceed the lesser of three times the slab thickness or 300mm.
Clauses 9.3.8.2.4 and 8.8.1 specify that distribution reinforcement shall satisfy the requirements for
temperature and shrinkage but with a ratio of reinforcement area to gross cross sectional area of
0.7/fy but equal to or greater than 0.0014.
When designing deck slabs that utilise partial depth precast decking (taken as 100mm thickness), the
transverse reinforcement can be conservatively designed as being located in the remaining 150mm
concrete topping. When designing for the longitudinal reinforcement, the total deck thickness of 250mm
is used. Note that all such slabs composing of partial depth precast decking with an in situ topping shall
be designed in accordance with section 18 of NZS 3101, to ensure composite action between the
components through detailing the appropriate reinforcement.
Positioning of the reinforcement in the slab is dependent on the specified cover thickness to the outer layer,
in accordance with section 3 of NZS 3101. Specifying the correct concrete cover is crucial to ensuring the
performance of the slab over the design life of the bridge structure. The cover is dependent on the exposure
classification according to its location (given in table 3.1 of NZS 3101) and the compressive strength of the
concrete. Table 3.7 of NZS 3101 provides the different concrete covers based on these factors. Bridge deck
slabs are usually specified with 40MPa concrete that has a minimum cover ranging between 35mm and
50mm for the different exposure classifications of inland (A2), coastal perimeter (B1) and coastal frontage
(B2) respectively. Note that the correct curing of the concrete is another important factor in ensuring the
performance of the concrete member. See section 8.1 for further details.
7.10.2.1 Slabs in multi-girder decks
In a multi-girder deck, the greatest bending moments in the slab due to the traffic loading are transverse;
both sagging and hogging moments are generated (sagging is usually greater). Global bending due to the
differential deflection of the main girders also contributes to transverse bending in the slab.
Consequently, the transverse reinforcement is taken as the principal reinforcement and is usually placed
as the outer layer in multi-girder bridge slabs.
7.10.2.2 Slabs in ladder decks
In a ladder deck bridge, the greatest bending moments in the slab due to traffic loading occur in the
longitudinal direction: both hogging and sagging moments are generated. With long cross girders the
deflection under the most heavily loaded girder does make a significant contribution to the total sagging
moment.
Transverse moments (due to the local effect of wheel loads) are smaller and mainly sagging in nature the
exception is over the length of the cantilever and immediately inboard of the main girders.
Consequently, the longitudinal reinforcement is taken as the principal reinforcement, usually placed as the
outer layer in ladder deck bridge slabs. However, this reduces the lever arm for the transverse moment,
which is a disadvantage for the transverse reinforcement in the cantilever, but the penalty is normally
accepted.
Slabs in ladder deck bridges are designed as two-way slabs, due to the above moment actions and the
punching shear requirements. This is clarified in section 7.10.3 below.
106
Detailed design
7.10.2.3 Effect of the axial load

It was noted that the effect of axial loads on the bending resistance of the composite section is not
covered in either NZS 3101 or AS 5100.
Where the slab is in tension, ie in the hogging moment region, then additional reinforcement may be
required at the top of the slab that is designed to carry the tensile force acting on the slab at that point
(See clause 6.1.4 of AS 5100.6 for additional guidance on minimum reinforcement). When the slab is in
compression, ie in the sagging moment region, the compression forces are typically resisted by the
concrete which is assumed to be fully effective as stated in clause 7.4.2.1 of NZS 3101.
However, where a ladder deck slab is in axial compression, a question then arises about its slenderness,
over the lengths between cross girders. There is no guidance available on the maximum spacing between
the main girders where the slab is considered to be fully effective in compression. Clause 7.4.2.1 of
NZS 3101 states that it can either be assumed to be fully effective or a complete stress-strain analysis
must be undertaken. To address this issue, Iles (2010, section 6.3.2) states that if the spacing of the main
girders does not exceed about 30 times the thickness of the slab, the slab may be considered as fully
effective in compression (it acts as a plate supported on four sides). For a wider spacing of the main
girders, the slab tends to act as a wide slender strut in compression and its slenderness reduces its axial
resistance, because second order effects are introduced. However, the slab is not usually fully utilised in
compression and the reduction is acceptable.
7.10.3 Design for shear

The basic requirement for design for flexure in slabs is given in clause 7.5:

(Equation 7.46)
Where the design shear action (V*) derived from the ULS loads is less than or equal to the factored
nominal shear strength (Vn) of that section of the slab.
The requirements for the design for shear of two-way slabs are given in clause 12.7.2. In that clause it is
noted that the shear strength is governed by the more severe of beam action and two-way action: for a
deck slab supporting wheel loads, the two-way action will be more severe, due to the concentrated nature
of the loading. However, it is recommended that both cases are considered when designing for shear, as
the slab shear capacity is reduced when the slab is in tension, which may affect the slab overall capacity.
Clause 12.7.2 refers to clause 12.7.4 for determination of shear strength and notes that this is based on
the general requirements in clause 7.5.3; the basic requirement is expressed as:
Where:
=
+
(Equation 7.47)
Vn = nominal shear strength of the section

Vs = nominal shear strength of shear reinforcement
Vc = nominal shear strength from the concrete.
The values of the minimum shear strength of shear reinforcing and concrete are given in the
accompanying clauses 12.7.3 and 12.7.4.
Clause 12.7.3.4 limits the maximum nominal shear stress for punching shear on any part of the perimeter
. When the nominal shear stress (v ) on any part of the critical perimeter exceeds the critical
to 0.5
n
c
around the complete perimeter according to clause 12.7.3.5.
shear stress (vc), then (vc) is reduced to

6
107
Where the applied shear stress exceeds that provided by the concrete, shear reinforcement in the form of
anchored bars, wires, single or multiple leg stirrups can be provided in slabs in accordance with clause
12.7.4.1, where the effective depth of the slab is greater than or equal to:
150mm and
16 times the diameter of the shear reinforcement.
An additional and important requirement in New Zealand is related to the minimum area of shear
reinforcement that is given in clause 9.3.9.4.13. This clause states that when highly repetitive loads (ie
vehicle loading) are acting on the slab, shear reinforcement is required. The minimum area of shear
reinforcement is given in clause 9.3.9.4.15, which is:
1
16
(Equation 7.48)
The spacing limits for shear reinforcement are given in clause 9.3.9.4.12.
Note that in the hogging moment region, the deck slab is in longitudinal tension, which reduces the
concrete shear capacity, thereby the nominal shear strength of concrete (Vc) is taken as zero.
7.11 Fatigue design

7.11.1 Introduction to fatigue
Fatigue is one of the most important design checks that must be undertaken when designing a steelconcrete composite bridge. Fatigue cracks grow from very tiny imperfections when there is a fluctuation of
stress across the imperfection that tends to open it. However, this does not mean that only an overall
tensile stress in the part is relevant to crack growth, because any stress range will tend to open and close
the initiating imperfection; the reference stress level for classification is therefore the stress range, which
in this case is taken as the maximum peak-to-peak stress level due to the application of the cyclic load.
Fatigue failure of steel depends on the propagation of these imperfections (usually in the welds) in regions
that are subject to fluctuating stress. The fatigue life therefore depends on the detail category (which
accounts for the size of the initial imperfection and, to some extent, the local stress concentration) and on
the range of the stress variation (number and magnitude of cycles).
Fatigue assessment is therefore a two-stage process:
1
Determination of the fatigue design spectrum the number and magnitude of the cycles of stress
variation as a result of the frequent loading on the bridge.
Determination of the fatigue life of particular details for the given fatigue spectrum.
7.11.2 Fatigue design spectrum

The basic requirements for determining the design spectrum are given in clause 13.3. It requires that
stresses are calculated using an elastic analysis of the structure (which implicitly requires the effects of
shear lag to be taken into account). Clause 13.3.2 states that the stress spectrum shall be obtained by a
stress cycle counting method.
In general, a bridge will be subject to numerous different stress ranges, each with its number of cycles
over the design life. Fatigue assessment would then need to take account of the damage caused by each
range to obtain a cumulative assessment for the entire loading. This is a rather time consuming and
difficult exercise and for that reason a simplified fatigue loading model is normally used. The choice of a
108
Detailed design
suitable loading model for New Zealand traffic is discussed in section 6.4.2. This simplified model is used
to determine an equivalent constant stress range for a specified number of cycles.
7.11.3 Fatigue strength

Fatigue strength is conventionally expressed in relation to the endurance (number of cycles to failure) of a
particular structural detail subject to constant amplitude stress. The relationship between fatigue strength
and number of stress cycles is referred to as an S-N curve (plotted on a log-log scale); examples for normal
stress and shear stress are given in clause 13.6 for a range of detail categories. The detail categories are
discussed below. Note, however, that the fatigue strength for transverse fillet or butt welded connections
needs to be reduced or corrected when the connected plate thickness exceeds 25mm (see clause 13.1.7
and section 7.11.5).
7.11.4 Classification
Fatigue detail classifications relate to the potential imperfections at welds, holes or other discontinuities,
and their relationship to the stress direction. The greater the imperfection, the lower the fatigue strength
is for a given number of cycles. The complete range of classifications is shown in table 13.5.1 of
AS 5100.6. This table shows a range of typical details in detail category classification group 1 to 4, which
includes welds and shear connectors.
Generally, lower category details are introduced by making attachments to the steelwork component.
Fatigue assessment therefore needs to be carried out chiefly in regions of significant variations of stress
and at the locations of the attachments. Typical locations requiring detailed assessment are: webs and
flanges over internal supports; at the attachment of web stiffeners, flange or web reinforcing plates; all
connections in transverse bracing and at splices (welded or bolted).
The attachment of web stiffeners or other elements not carrying load in the stressed direction usually
produces a category 71 or 80 detail. Reinforcing plates and bearing plates welded to the underside of the
flange usually introduce a category 45 or 50 detail. Shear stud connectors introduce a category 80 detail
in the plate to which they are attached (see section 7.7.6 for further details). Preloaded bolted splices
introduce category 140 details for double covers or single covers. Transverse butt welds create category
50, 80, 90 or 112 details and a size effect reduction factor applies as specified in table 13.5.1 of
AS 5100.6.
The reliability of the classification of a particular detail depends on a presumption about the quality of
workmanship. To ensure that the workmanship will be appropriate to the fatigue life and detail class
assumed by the designer, clause 13.1.5 states that the weld details given in table 13.5.1(B) and 13.5.1(D)
for detail category 112 shall conform with category SP (structural purpose) to AS/NZS 1554.1 (SNZ 2004c).
While the weld details in table 13.5.1(B) for detail category 125 shall have a weld quality conforming to
that defined in AS/NZS 1554.5 (SNZ 2004d). Workmanship levels, which are specified in AS/NZS 1554 (SNZ
2004) and NZS 3404.1 (SNZ 1997b), relate the level of inspection and the acceptance criteria for
imperfections to the minimum fatigue class required by the designer.
7.11.5 Fatigue assessment

Clause 13.7 sets out how detail categories in a particular structure are to be assessed for acceptability for
fatigue design. In making this assessment, the design fatigue capacity is determined by applying a
capacity reduction factor (), as given by clause 13.1.6. Although the value of (), is generally to be taken
as 1.0, for non-redundant load paths = 0.7 is used. It should be noted that the non-redundant load
path
109
is taken as the path in a member where failure would lead to the instability resulting in partial or complete
collapse of the structure.
Clause 13.7.1 gives an exemption from the need for detailed fatigue assessment for situations where the
stress range is less than the corrected detail category fatigue strength at the constant amplitude fatigue
limit. This is expressed by:
for normal stresses:

5
(Equation 7.49)
for shear stresses:
Where:
5(Equation 7.50)
f*n
= design stress range for normal stresses
fnc
= corrected detail category fatigue strength at constant amplitude fatigue limit for normal stresses
f*s
= design stress range for shear stresses
f5sc
= corrected detail category fatigue strength at constant amplitude fatigue limit for shear stresses.
Clause 13.1.7 states that the thickness correction factor (tf) that is needed to determine (f5nc) and (f5sc)
shall be taken as 1.0, except for a transverse fillet or butt-welded connection with a plate thickness (tp)
greater than 25mm. In that case the uncorrected fatigue strength for the different detail categories is
multiplied by the following expression, given in equation 13.1.7(1) of AS 5100.6.
0.25
= 25
(Equation 7.51)
Clause 13.7.2 gives the requirements for assessing the detail category in the constant stress range, or
when the design stress range (f*n) or (f*s), is calculated using the simplified fatigue loading (see section
6.4.2) and the effective number of fatigue cycles (n). Then the detail being assessed for normal and/or
shear stresses is deemed to comply if it satisfies the expression given in this clause.
Clause 13.7.3 gives requirements for assessment for variable stress range. This method is used where the
loading such as vehicle or wind loadings acting on a bridge are variable but the magnitude and
frequencies of each range are known. The variable stress range, is based on the damage accumulation
method (commonly known as Miners summation), as the design stresses and their damage is
accumulated when assessing whether the detail complies with the clause requirements. The variable stress
range must satisfy the following expression:
5
5106 (
3)3
1.0
(Equation 7.52)
5106 (
3)5
To determine the variable stress range on a multilane bridge, assuming two heavy vehicles are acting on
the bridge, the stress range (f*) for each heavy vehicle is first determined as an independent load, as
described in section 6.4.2. The stress range for each vehicle is then compared with the constant stress
r
range fatigue limit (f3) for the detail category (f n) being considered given in figure 13.6.1 of AS 5100.6. If
the stress range for the vehicle is less than the constant stress range fatigue limit, then the left hand part
of the equation is used. Conversely if the stress range of the vehicle is greater than the constant stress
range then the right hand part of the equation is used. In addition to the above requirements, an
accompanying lane factor shall be added to the second lane vehicle in accordance to clause 6.6 of
110
Detailed design
AS 5100.2. The summation of stresses for both vehicles must be less than or equal to (1.0) to satisfy this
expression.
Clause 13.7.4 gives a more general version of the variable stress range method, expressed in a manner
suited to assessment of existing structures or when the bridge design life is other than 100 years.
If designers find that fatigue assessment has an unacceptable outcome, they may choose to specify a
better class of detail, to modify the design to reduce the stress range or, if the simplified model has been
used, to re-assess by the variable stress method.
Finally, as briefly discussed in section 4.2.3, clause 8.4.2.2 of NZS 3101 prohibits the use of bent
reinforcement bars less than 20 times the bar diameter for concrete members subject to frequently
repetitive loads. This is applicable to vehicle loads that expose the reinforcing bars to potentially high
levels of fatigue. The fatigue design requirements and the permissible stress range for reinforcement bar
is given in clause 2.5.2 of NZS 3101. While the permissible stress range for concrete is given in clause
2.5.2 of AS 5100.5.
7.12 Selection of steel sub-grade

The following guidance is based on section 2 of NZS 3404.1 (SNZ 2009), as it contains New Zealand
specific steel selection requirements, which include temperature and seismic requirements.
All parts of structural steelwork are required to have adequate notch toughness, to avoid the possibility of
brittle fracture. Brittle fracture can initiate from a stress concentration when loading is applied suddenly, if
the material is not sufficiently tough. The degree of toughness required is expressed as a Charpy impact
value (determined from a test carried out on a sample of material) and the requirement depends on the
thickness of the material, its minimum temperature in service, seismic requirements, the stress level and
rate of loading.
The design requirements for bridges are given in clause 2.2.5 and table 5 of NZS 3404.1, in terms of the
steel type and its limiting thickness for the given conditions and specified material. If steel type and its
thickness comply with the requirements, then it is considered to be sufficiently tough.
The most obvious condition that needs to be considered is the lowest temperature that the steel will
experience. The basic design temperature is specified in clause 2.6.3 of NZS 3404.1 as the lowest one-day
mean ambient temperature (LODMAT), which depends on the bridge location around New Zealand.
According to figure 1 of that standard, the lowest temperatures in New Zealand range from -10C in the
South Island to 7.5C in Northland. However, for bridges, site-specific low temperatures are required,
especially in the alpine areas in the South Island. In some cases the basic design temperature is reduced by
5C (see clause 2.6.3.2 of NZS 3404.1).
Clause 2.2.5 of NZS 3404.1 gives the steel type chosen for bridges where they are fracture critical
members (FCM). These are components in the bridge, the failure of which would result in the collapse of
the bridge. FCM are specifically required for all railway bridges or highway bridges with significant
overloads and with high fatigue loading.
In most cases, non-FCM bridges will commonly use steel type 4, either grade 350 to AS/NZS 3679.1 (SNZ
2010a) or WR350 to AS/NZS 3679.2 (SNZ 2010b). Alternatively, for FCM bridges steel type 5 will
commonly be used, either grade 350L0 to AS/NZS 3679.1 or WR350L0 to AS/NZS 3679.2, both of which
have a Charpy impact value of 27J at 0C.
111
Other steel grades with higher Charpy impact, such as L15 grade with 27J at -15C or the seismic grade S0
with 70J at 0C, are available. However, the seismic grade is rarely used in bridges, unless it is used in the
substructure (ie piers).
Table 5 of NZS 3404.1 gives the permissible service temperatures according to the steel type and
thickness, which must be considered by the designers when specifying the steel type for the bridge.
112
Construction considerations
8.1
Steel fabrication and concrete construction
The design rules in AS 5100.6 (SA 2004) are valid for steel components that are fabricated in accordance
to NZS 3404.1 (SNZ 2009), which includes specification of materials and of fabrication workmanship.
All concrete components that are designed to NZS 3101 (SNZ 2006) shall be constructed in accordance to
NZS 3109 (1997a). Note that proper curing of the concrete is crucial to the structural and durability
performance of the concrete member. Therefore, the designer should ensure that the contractor meets the
minimum requirements listed in NZS 3109, which can be achieved by providing a financial incentive by
stipulating the curing of the concrete as a separate item on the project contractual documents.
8.2
Construction sequence
The construction sequence that most commonly needs to be evaluated for a composite bridge is
completion of the substructures, up to bearing level, erection of the structural steelwork (piece by piece),
provision of formwork and casting the deck slab, and finally completion of the surfacing and fixtures such
as barriers and drainage. Each construction stage needs to be analysed; a series of models of the partly
completed structure is required for each stage. Where construction methods such as launching and
transportation of the part-completed structure are used, the local effects at temporary support positions
need to be evaluated.
Where it is not practicable to cast the full length of deck at once, the series of analytical models must
represent the development of the composite structure as the portions of slab are cast.
The results of the analytical model for the chosen construction sequence will then be used to design the
structural components for that construction stage, following the guidance given in chapter 7.
8.2.1
Girder erection
Girder lengths are usually chosen to suit transportation (see section 4.2.1 and Guide to heavy vehicle
management (NZTA 2006)), although the weight of individual pieces may limit the sizes where crane
access is restricted. Strength verification at this stage is unlikely to require detailed evaluation but stability
and buckling resistance do require careful consideration, particularly before bracing or cross girders are
fully installed.
8.2.2
Bracing
Bracing of the steelwork in the bare steel and partly complete stages is a key to the effective performance
of the main girders. Several bracing schemes may need to be evaluated.
8.2.3
Slab construction
Although deck slabs have traditionally been cast on temporary timber false work, the use of permanent
formwork notably partial depth precast decking that forms part of the final slab is now very common
(section 2.3). Timber false work is often supported off the bottom flanges of the girders; precast
permanent formwork sits on the top flanges and thus needs to be considered as a destabilising load.
Whichever type of formwork, the weight at the wet concrete stage imposes quite high stresses in the top
113
flanges of the girders. Their strength and stability at this stage require a detailed evaluation of the
progressive changes in structural behaviour as load is added.
The weight of the concrete cantilevers needs particular attention, because of the moment (about the
longitudinal axis) that is imposed on the outer girder. See further comment in section 8.3.1.
8.2.4
Patch loading on webs
For girders that are erected by launching, reactions under the girder as it is progressively launched impose
local patch loading on the unstiffened portions of the main girder webs. The web will need to be checked
for the effects of combined stresses and for buckling. See section 7.4 for details of web design.
8.3
Cantilever edge slabs
8.3.1
Loading from cantilevers
The use of precast units, especially partial depth precast decking, is the current popular method of
cantilever construction.
The moment due to weight of cantilevered false work and wet concrete is transferred to the main girder as
a couple of horizontal forces at top and bottom flange levels; these forces cause horizontal bending of the
flanges between restraint positions. This is in effect warping torsion, rather than St Venant torsion. In
ladder deck bridges, the effects of warping are modest, because the cross girders provide restraint at
close regular intervals; in multi girder decks the restraint positions are further apart and the effects are
greater. Deflection at the restraint positions (due to the bending of cross girders or the vertical
displacement of the main girders due to the eccentric moment) adds twisting effects. Warping stresses,
distortional displacements and twists all need to be determined.
Although the warping stresses (transverse bending stresses) in the top flanges are locked in once the
concrete hardens, it is not necessary to include these effects for the in-service condition because at ULS
they will redistribute and at SLS any relaxation would be unlikely to lead to any noticeable permanent
deformation.
The alternative method of constructing cantilevers is to add full thickness precast units once the central
portion of the deck slab has been completed. Support for these units can be from overhead temporary
frames on the deck and although the weight causes twist (because of differential deflection of the main
girders); there are no warping effects in the main girders.
8.3.2
Design of cantilever edge slabs
For multi-girder bridges, cantilever edge slabs are usually the last part of the deck slab to be concreted,
in order to achieve a good alignment along this very visible feature. Their contribution to structural
behaviour of the cantilevers cannot therefore be relied upon until a late stage during construction. The
edge slabs of ladder deck bridges are usually concreted at the same time as the main deck slab.
8.4
Allowance for permanent deformations
The deflections under unfactored dead and superimposed loads should be calculated to enable the girders
to be pre-cambered. This information should be produced by the designer and a breakdown of the effects
of the various actions included on the drawings. Where staged construction has been presumed, the
sequence should be stated on the drawings.
114
A residual hogging profile is often specified, for aesthetic reasons, even when not needed to meet a
clearance requirement at SLS. For the calculation of deflections of composite sections, it is necessary to
assume an age at first loading, so that the appropriate parameters for concrete can be determined. The
steelwork should normally be pre-cambered to offset the predicted deflection at the end of construction.
115
Detailing considerations
9.1
Geometric configurations
The designer should clearly and unambiguously define the geometric configuration of all the main
structural elements of the bridge. The following comments relate to certain specific issues of good
practice for multi-girder and ladder deck bridges.
9.1.1
9.1.1.1
Road camber and crossfall
Usually the top flanges of the main girders are square to the vertical webs. The relationship with a deck
slab that follows the road camber or a transverse crossfall then needs to be considered carefully. There
are four main options:
1
Keep the slab soffit level and the thickness uniform; the crossfall is achieved by varying the thickness
of the surfacing.
Keep the slab soffit level and vary the slab thickness, so that the top surface follows the required
crossfall.
Slope the slab soffit between the edges of the girder flanges; the top surface follows the required
crossfall and the slab thickness varies across the width between girders.
Provide small haunches above the girder flanges and use a uniform thickness slab, following the
crossfall.
The first and second options are only appropriate for two- or three-lane bridges with no superelevation, where the weight penalty is modest. Variation of surfacing thickness is preferred to variation
of slab thickness, for economy in construction. The first three options all suit the use of permanent
formwork; option 3 is perhaps the most common and is the arrangement shown in figure 2.1. Options 1
and 2 are illustrated in figure 9.1.
Figure 9.1
Two options for dealing with road camber in single carriageway multi-girder bridges
3%
3%
3%
3%
116
Option 4 is suited to the use of timber temporary formwork and, until the increased use of permanent
formwork, was the most common arrangement. A typical arrangement for option 4 is shown in figure 9.2.
The haunches can be formed relatively easily and the resulting slab is of uniform thickness. The use of
haunches also makes it easier to accommodate any unintended differences in relative level between
adjacent girders that are found after steelwork erection.
Figure 9.2
Use of haunches with a multi-girder slab constructed on temporary formwork
3%
3%
For all four options, designers usually choose the same girder depth for all girders. For options 3 and 4,
the girders are at slightly different levels.
Where the deck is wider, or where there is super-elevation (uniform gradient across the full width of the
carriageway), the arrangements are then similar to those illustrated in figure 9.3. See further comment on
the effect of a varying slab thickness in section 9.4.
Figure 9.3
Two options for dealing with deck slab with super-elevation in multi-girder bridges
Min. thickness of
deck slab
Permanent
formwork
Uniform slab
thickness
Haunch
9.1.1.2
Bracing planes
Planes of bracing are usually square to the top flange, rather than vertical. As noted in section 2.1.3, the
planes are usually square to the main girders in plan.
117
9.1.2
Ladder decks
9.1.2.1
Plan layout
The position or spacing of cross girders should be defined in relation to their centrelines (ie the midthickness of the web). This is particularly important where a lapped connection is used, to avoid confusion
if the position were defined to one face of the web or to the centreline of the stiffener to which the cross
girder is attached. It also helps to ensure that the fabricator and the supplier of the formwork are working
to the same dimensions.
On curved bridges, the configuration of the main girders can be arranged to follow the curvature of the
roadway, to a uniform radius, to a spiral or to a mixture of straights and curves. Cross girders should be
arranged radially to the defining curve (normally the centreline of the road).
9.1.2.2
Road camber, crossfall and longitudinal gradient
Transversely, the top flange of the cross girder will usually follow the camber of the roadway or the superelevation of the roadway. The alignment of the flange in relation to the flanges of the main girders needs
to be considered: if the main girder flanges are horizontal (across the bridge), variations in slab thickness
and the consequences of any variation in width of main girder flange need to be taken into account. For
the usual crossfall (3%), or a modest crossfall to provide super-elevation, the top flanges of the cross
girders can be aligned as shown in figure 9.4; the small step at the edge of the main girder flange does
not introduce construction difficulties, although care will be needed in sealing between permanent
formwork and the main girders.
Figure 9.4
Alignment of flanges
Longitudinally, the top flanges of the cross girders should be aligned with the longitudinal profile of the
main girders, which follow the longitudinal road profile; this will maintain uniform slab thickness along
the bridge. To avoid complexity in fabrication and the need for a different cross sectional geometry of
every cross girder, each cross girder should be detailed with parallel flanges square to its web and the
cross girder should then be connected with the web square to the main girder top flange. This means that
where the cross girders and the main girder web stiffeners connect will, in general, not be truly vertical
and their inclination will vary along the bridge. This does not cause difficulty for the fabricator.
There is no explicit requirement for bearing stiffeners at intermediate or end supports to be truly vertical;
although designers usually prefer to detail them to be vertical under dead load (it is visually better for the
bearing stiffeners on the outer faces to be vertical). Where there are integral crossheads, it is structurally
better to make their webs vertical, to minimise the twisting effects from the bearing reaction. If a pier
cross girder, diaphragm or integral crosshead is detailed with the web vertical, its top flange should still
follow the longitudinal profile of the main girders and the flange will then be slightly tilted relative to the
web, as shown in figure 9.5.
118
Figure 9.5
9.1.3
Alignment of diaphragm girder flange when its web is vertical
Allowances for permanent deformation
The steelwork is normally fabricated with allowances for permanent deformation due to the weight of the
complete structure (see section 8.4) and for cutting and welding during fabrication. The fabricator can
deal with these allowances in determining the shape of all the elements that are to be cut from steel plate,
provided the designer advises the allowances for the intended construction sequence (see section 8.4).
However, there are still some questions that arise with a composite structure, such as at what stage are
the webs at supports to be truly vertical: under the weight of steelwork alone or after completion?
Although designers might wish to select the latter option, it is well known that it is very difficult to predict
rotations of main girders (about their longitudinal axes) at supports, particularly for skew bridges. It is
therefore commonly arranged that the webs are fabricated to be vertical under the weight of bare
steelwork and the girders are designed for the out-of plumb that would result under the weight of
concrete and superimposed load (assuming that rotations occur as predicted).
Guidance on fabrication and erection tolerances is given in appendices G and H of AS 5100.6 (SA 2004),
which is similar to the guidance given in sections 3 and 4 of NZS 3404.1 (SNZ 2009).
9.2
Cross girder end connections
9.2.1
Intermediate cross girders
Intermediate cross girders (ie away from supports) are connected to web stiffeners by simple lapping of
the web plate onto the stiffener, as shown in figure 9.6. Both flanges are stopped short of the end of the
web; this is easily achieved with a fabricated girder. Double cover plate splice connections are
inappropriate because the same number of bolts is required as with a single lapped connection, and
additional cover plates have to be fabricated, making the detail more expensive. However, if rolled
sections are chosen for the cross girders, a double cover detail may be preferable (allowing the section to
be cut with a plane end) because of the cost of the work that would otherwise be needed in the cutting
back of flanges of a rolled section.
119
Figure 9.6
Figure 9.7
Lapped connection of intermediate cross girder
Details of lapped connection
In most cases, this simple lap detail, with no connection of the flanges, will be sufficient to transmit the
moments in the U-frame (see section 7.8.6). If the moments are greater than can be transmitted this way
(perhaps because a very shallow cross girder is chosen) a connection of the bottom flange similar to that
sometimes used for pier diaphragms (see figure 9.11) may be effective.
With a lap detail, the cross girders are longer (over the length of the web) than the clear gap between the
main girder flanges. Consequently, during erection they are lifted at a skew (in plan) so that they can be
lowered past the top flanges and then rotated and brought into lapping contact with the main girder
stiffeners (see figure 9.8). Ideally, plan rotation as shown should be possible with both the adjacent cross
girders in position. It may be necessary to keep the end of the cross girder web sufficiently clear of the
120
face of the main girder web (and this may require a wider web stiffener) so that the cross girder can be
erected in this manner even when the cross girders on both sides are already in place. The laps at the two
ends of the cross girder should be to the same face of the web, to avoid the risks of confusion and error in
setting out and installation.
Special attention should be given where there are jacking stiffeners between intermediate cross girders, or
small bays between cross girders. The cross girders could be trapped between stiffeners, preventing the
rotation illustrated above. The detailing may need to be adjusted in these areas, depending on the layout
and the construction sequence.
Lapped connections are less accommodating of deviations of the cross girder length (and of the layout of
the bolt holes) than a spliced connection. With modern fabrication techniques this should not be a
problem, although some designers have allowed for oversized holes in the design (the slip resistance is
reduced) in case reaming should prove necessary.
Figure 9.8
9.2.2
Schematic arrangement for erecting intermediate cross girders
Pier diaphragms
Lap type connections cannot readily be made at intermediate supports as, once the main girders are in
place at the required spacing, the pier diaphragm girder will foul on the jacking stiffeners as it is swung
into place. Hence, cross girders framing into the bearing stiffeners will normally be connected using
double cover plates to the web. See figures 9.9 and 9.10 for further details.
121
Figure 9.9
Spliced connection of pier diaphragm
Figure 9.10
Arrangement of double lap splice connection of pier diaphragm
To transfer larger moments at supports, a cover plate connection to a stub flange attached to the bearing
stiffener may be needed (see figure 9.11). A double cover plate connection should be provided.
Connection of the top flange should not be necessary. The cross girder should be less deep than the main
girder, to avoid any need to connect to the main girder bottom flange.
Figure 9.11
Elevation on splice connection of pier diaphragm with stub flange
122
9.2.3
Integral cross heads
Where the bridge is supported on bearings under an integral crosshead, the load on the connection is
clearly much greater and more bolts will be required. A typical connection arrangement is shown in figure
2.12 and some of the details are shown at larger scale in figure 9.12.
If longitudinal restraint is provided at an integral crosshead, a suitable load path (in terms of both
strength and stiffness) must be provided for horizontal restraint forces at a longitudinally restrained
bearing. This should normally be provided by designing for plan bending of the bottom flange of the
crosshead.
Figure 9.12
9.2.4
Connection of integral crosshead to main girder
Crosshead girders in multi-girder bridges
Where crosshead girders are used (such as shown in figure 2.5) bolted connections will usually be needed.
Alternatively, the crosshead and lengths of the two main girders which it supports can be fabricated as a
single H section (in plan). This can reduce the amount of site work making connections. The overall
dimensions of the H section are limited by transport restrictions.
9.3
Shear connection
Even though clause 6.6 references three types of shear connectors (see section 7.7), the most commonly
used shear connectors are the headed stud. Headed studs, also known as shear studs, are available as
15.9mm, 19.0mm and 22.2mm diameter, of which the 19mm diameter is the most commonly used. They
are readily available and can easily be welded using a special semi-automatic welding tool. Stud
dimensions and minimum spacing limits are given in clause 6.6.2.
Because studs are required to prevent separation, the undersides of the heads of the studs need to be a
minimum distance above the bottom layer of reinforcement. Requirements for haunched and unhaunched
configurations are shown in figure 6.6.5.2 of AS 5100.6 (SA 2004).
123
The use of 125mm or 150mm long stud connectors will ensure that the heads are well above all bottom
transverse reinforcement in most cases.
The stud longitudinal spacing shall not be greater than the minimum of 600mm, three times the slab
thickness or four times the height of the connector (which leads to the need for studs on cover plates on
the top flange, as shown in figure 7.18). They should also not be closer than 25mm to the edge of a flange
(50mm if the slab is haunched); larger edge distances are needed when precast permanent formwork is
used, to ensure secure seating of units.
9.4
Deck slab
The detailing issues related to deck slabs concern chiefly the location of reinforcement.
9.4.1.1
Multi-girder decks
In multi-girder decks, the transverse reinforcement is normally placed as the outer layers and the
longitudinal reinforcement is placed as the inner layers. Where precast permanent formwork is used
between the main girders, only the in situ lower transverse rebars are effective in transferring shear to the
main girders; the adequacy of these bars needs to be verified. The location of the upper longitudinal
rebars in the inner layer suits fixing of reinforcement, as they can sit on the top of the protruding hoops
of the precast units (although see note below about varying slab thickness).
Where permanent formwork is not horizontal, the thickness of the slab varies, as noted in figure 9.3. The
arrangement of the transverse reinforcement at a main girder is illustrated in figure 9.13. It is not usual
(or desirable for fatigue reasons) to crank the transverse bars in either the top or bottom mat, although
they will bend a little. In the top, the bars may lift off slightly from the plank reinforcement on the higher
side and the cover may be slightly greater. The bottom bars may also be slightly higher at the same
location.
Figure 9.13
Local cross section at main girder in multi-girder deck with precast permanent formwork
precast
Precastplank
plank
Section
Permanent formwork needs to be constrained by the positions of the studs at each end so that it cannot
displace and fall through, between the girders, during construction. The nominal bearing length for
precast decking is typically 75mm and studs should be no more than 25mm inside the nominal position of
the decking ends; this will ensure that the decking cannot be displaced along their length and then fall
through. A seating rubber strip is recommended to take into account any height differences in the decking
placement. The designer should also consider the positioning of the reinforcement and available flange
width to ensure the above dimensions are met.

Any protective treatment to the steelwork should be continued inward from the top edges of the flanges
to a minimum of 100mm. A sealant will also be required where permanent formwork is sloped, relative to
the flange and in all cases where the girder is of weathering steel.
124
9.4.1.2
Ladder decks
The transverse reinforcement is usually the inner layers in the slab and the higher position of the bottom
rebars (higher than the outer layer) needs to be recognised when considering their position relative to the
underside of the shear stud heads.
Where precast permanent formwork is used, the transverse reinforcements in the main body of the slab
are even higher. To achieve the necessary clearance below the head of the studs bars need to be cranked,
or additional U-bars provided, see figures 9.14 and 9.15. Alternatively, taller studs can be used.
Figure 9.14
Cranking of transverse reinforcement
Min 25mm
Precast plank
30mm crank
Cross girder
Figure 9.15
Alternative use of U-bars
precast
Precastplank
plank
cross
Crossgirder
girder
Consideration should also be given to ease of fixing the top mat of the slab reinforcement. Usually the
transverse bars would be detailed in the top layer as this gives maximum lever arm for the tension
reinforcement at the root of the deck cantilever. However for steel fixing it is easier to place the transverse
bars as the lower layer, directly onto the precast plank lattice (layer T2 in figure 9.16), and then place the
longitudinal bars in the top layer (ie layer T1 in figure 9.16).
As with multi-girder decks, the positions of the studs should be such that the precast units cannot fall
through during construction.
125
Figure 9.16
Positioning of top reinforcement in slab
T1
T2
precast
Precastplank
plank
Crossgirder
girder
cross
As for multi-girder decks, protective treatment should be returned along the top surface of the top
flanges edges by a minimum of 100mm and sealants used where appropriate.
9.5
Bearing specification
Bearing design and construction are covered by AS 5100.4 (SA 2004).

Bearing design is usually the responsibility of the bearing manufacturer; the bridge designer should
provide a specification for each bearing, listing the range of reaction forces and movements (translational
and rotational).
The most commonly used type of bearing for highway bridges in New Zealand are elastomeric bearings
(section 12 of AS 5100.4); consisting of a single unbounded layer or laminated elastomeric bearings. The
other is the pot bearing (section 13 of AS 5100.4); consisting of a disk of elastomer confined in a short
cylinder, onto which the reaction is transferred by a piston. Both types of bearings accommodate
moderate rotations in any direction but are relatively stiff vertically. If a sliding surface is provided within
the bearing, translational movement can be accommodated; freedom can be provided in any direction, or
guides may be provided to confine movement to one direction. Without a sliding surface, full translational
restraint is provided.
Displacements due to permanent actions in heavily skewed decks may include large rotations (about each
main girder axis) at both intermediate and end supports, and thus large transverse displacements at the
bottom flange level.
This is particularly true for ladder deck bridges. These effects are a function of the plan geometry of the
deck and are related to the magnitude of the dead load precamber required; they cannot be avoided. Due
allowance for these rotations should be made in the design of the bearings (include the rotations in the
bearing schedule) and the detailing of tapered plates to which the bearings are attached.
The increased flexibility of the cross section in decks with shallow cross-girders can lead to significant
splaying of the main girders under both permanent and variable actions, causing relatively large
transverse translations in the bearings.
The use of deep cross-girders at intermediate supports and concrete diaphragm beams at end supports
will greatly reduce these movements.
As well as specifying the bearing, the designer should consider the requirements for attaching the
bearing, both to the steelwork and to the substructure. The attachment or anchorage, usually with a
126
tapered plate between the flange soffit and the bearing, should be designed for both the vertical and
horizontal forces involved (see section 10 of AS 5100.4). The alignment of the bearings and the
identification of the direction of the principal movement should be made clear on the drawings.
9.6
Deck joints
Deck joints are usually designed and supplied by specialist manufacturers. The bridge designer is required to
provide a specification, giving the displacements and design actions on the joint. Guidance on the design
requirements and specification of deck joints is given in section 17 of AS 5100.4, with additional guidance
and modifications outlined in section 4.7 of the Bridge manual, which the designer shall design to.
127
10
Design of durable composite structures
Durability design of steel and concrete must be considered by designers and is an important aspect of a
successful cost-effective design of any bridge structure.
10.1 Durability design of steel and concrete

To complement the numerous benefits of steel-concrete composite construction and extend its life,
suitable protective measures must be taken to ensure structural performance over its design life. This is
especially the case for the wide range of corrosion environments for either steel or concrete in
New Zealand.
New Zealand is a long, thin mountainous country lying in the prevailing westerly wind belt of the Southern
Hemisphere. Although surrounded by sea, its two nearest significant land masses are one of the worlds
hottest continents, Australia, and the worlds coldest, Antarctica. All these conditions give New Zealand a
diverse climate and very wide range of corrosion conditions, which if not designed for properly, will
prohibit steel and concrete from reaching their full sustainability potential by premature failure. The
following sections provide a summary of the available recommended publications that will provide
information to design engineers on different durability issues and their solutions.
10.1.1 Durability design of steel

Any steel structure exposed to a corrosive environment must be designed to provide optimum long-term
performance with a minimal level of normal maintenance. Durability design will require either the use of
self-protecting stainless or weathering steel or conventional carbon steel with a corrosion protection
system utilising a corrosion protection coating. When conducting a durability design, the bridge designer
is recommended to determine the optimum solution, ie one that will achieve the most economic time to
first maintenance based on the structures performance and aesthetic requirements, design life and
location. The optimally designed structure, whether coated or uncoated, will minimise the initial material
and energy inputs, provide cost savings from reduced future maintenance, provide health and safety
benefits and, for coated structures, less on-site debris to be contained and disposed of.
HERA report R4-133 New Zealand steelwork corrosion and coatings guide (El Sarraf and Clifton 2011),
which is used in conjunction with the joint Australian/New Zealand standard AS/NZS 2312 (SNZ 2002)
Guide to the protection of structural steel against atmospheric corrosion by the use of protective coatings
and the New Zealand standard NZS 3404.1 (SNZ 2009) Steel structures standard, part 1. All three provide
guidance to allow an appropriate and cost-effective coating system for any type of structural steelwork to
be selected and then specified in a generic manner. They cover the different types of protective coatings
available, the calculation of design corrosion rates for any steel surface, interior or exterior, above or
below ground, as well as coating inspection and maintenance, among other topics. The calculation
method, outlined in section 10.2.2, is necessary if coatings are to be appropriately specified for the
specific range of environments to be encountered.
Weathering steel provides an alternative to the use of protective coatings especially for bridges. HERA
report R4-97 New Zealand weathering steel guide for bridges (El Sarraf and Clifton 2005) provides the
necessary guidance to ensure that dependable performance is realised for applications of weathering steel
for New Zealand bridges. It covers the limitations on the use of weathering steel, design issues such as
standards and detailing, construction issues such as bolting, welding and handling, as well as what to look
for when inspecting and maintaining a weathering steel bridge.
128
10
These publications provide improved information to design engineers on different durability issues and
their solutions. By specifying a protective coating system or utilising the benefits of weathering steel a
more sustainable and durable solution can be found.
10.1.2 Durability design of concrete

Concrete structures (and elements in a composite bridge) are robust. However, this robustness, even with
increased concrete quality and cover to reinforcement, will not compensate for gross errors in design or
construction. Design, detailing, specification, execution and maintenance all influence the durability of a
structure, regardless of the materials used in its construction.
Section 3 of NZS 3101 (SNZ 2006) provides comprehensive guidance on how to select the appropriate
concrete quality in relation to the cover to reinforcement to provide a structure that is required to be
durable in relation to the identified exposure classes throughout its design life. To achieve a durable
concrete structure, other aspects of the process of design, specification and construction are equally
important and should not be overlooked: in particular, achieving the minimum cover, attention to
detailing, crack width control, care during the execution of the works and the correct curing of the
concrete. These requirements are given in NZS 3109 (SNZ 1997a).
Therefore, for any concrete structure (such as the piles and abutments) and concrete elements (such as
the concrete decking in a composite bridge), these components will be durable for their intended design
life, if made from properly compacted concrete which is in compliance with the compressive strength and
other specified requirements, and in which the achieved cover to reinforcement meets the minimum levels
specified.
10.2 An example of the durability design of steel

10.2.1 Background information
10.2.1.1 General steps to determining an appropriate coatings system using AS/NZS 2312
The procedure outlined in El Sarraf and Clifton (2011) and AS/NZS 2312 (SNZ 2002) to determine an
appropriate coatings system is as follows:
1
Determine the design service life of the element being coated.
Determine the site-specific corrosivity category (ACC), which is derived from the first-year steel
corrosion rate. This rate is determined from the combination of:
a
macroclimate, plus
microclimate.
Determine the time to first (major) maintenance required for the coatings system.
Select an appropriate corrosion protection system to meet the environmental requirements of 2 and 3
stated above based on cost, performance and any owner-specified factors such as colour and
appearance.
In accordance with the Bridge manual, the design service life for bridges is 100 years, which is much
longer than the 50-year design life typically used for buildings. Note that the design life of the structure is
not usually the same as its durability rating, ie the years to first major maintenance. This point is made in
clause 1.6 of AS/NZS 2312 and clause C5.1.1 of NZS 3404.1 (SNZ 2009), where it is noted that the
protection offered by the coatings systems is usually shorter than the design service life of the structure,
129
which means due consideration must be given to its maintenance or renewal requirements at the planning
and design stage. It is only when components of the structure are not accessible for maintenance after
assembly that the corrosion protection system must remain effective for the design service life of the
structure. This important distinction must be recognised by designers and specifiers.
The process of selecting an appropriate coatings system is always site specific, and will typically be
surface specific where microclimate effects are important and different surfaces have different exposures.
10.2.1.2 Determining the first-year corrosion rate
Macroclimate
An important part of this guide is the maps of first-year carbon steel corrosion rate in New Zealand. These
maps are based on equations developed by (Hyland and Enzensberger 1998), using BRANZ corrosion data
published by (Duncan and Cordner 1991) and climate data from the National Institute of Water and
Atmospheric Research Ltd. The equations were determined as a function of the following variables which
are needed to determine the macroclimate:
distance from seacoast (0.5, 5, 20km)
average annual daily temperature
9am time of wetness (RH 80%)
annual rainfall
upper bound results used.
The equations were used to produce the macroclimate corrosion rate maps which greatly simplified the
process of determining the atmospheric corrosivity category, as seen in figure 10.1, which provides the
North Island first-year corrosion rates. The corrosion maps have recently been updated and simplified in
NZS 3404.1.
Microclimate
The other factor needed to determine the corrosion rate is the microclimate effects on the steel surface.
These depend on whether the steel surface is shaded, in a wet location, and whether the steel is in contact
with timber or concrete. The most significant microclimate effect is if the steel surface is sheltered from
rain washing but exposed to the windblown marine salts as this greatly influences the corrosion rate,
depending on the distance from the sea and its position in relation to the prevailing wind.
Each one of these factors will affect the corrosion rate by multiplying or adding to the macroclimate
corrosion rate determined above. Figure 10.1 demonstrates the effects of the microclimate on determining
the atmospheric corrosivity category.
130
10
Figure 10.1
Detailed (left) and simplified (right) North Island first-year carbon steel macroclimate corrosion rates
10.2.2 Waikato River bridge example

The following example is given to demonstrate the process of determining the life cycle costing
methodology and the differences between a standard durability design and a sustainable durability design.
A steel bridge is to be built for the NZTA, which is located on the Waikato River in Hamilton, 40km from
the sea. Section 10.2.3 starts with the methodology of determining the actual atmospheric corrosion
category that will be used to specify a coating system and calculate the corrosion rate of weathering steel.
10.2.3 Determine the atmospheric corrosion category

10.2.3.1 Determine the design service life of the element being coated
The bridge is to have a design life of 100 years as stated in the Bridge manual.
10.2.3.2 Determine the site-specific corrosivity category (ACC), which is derived from the first-year
steel corrosion rate
This rate is determined from the combination of the macroclimate and microclimate effects.
Site macroclimate effect:

The site macroclimate effect is determined from the North Island first-year carbon steel macroclimate
corrosion rate found in El Sarraf and Clifton (2011, appendix A.1), as 20 m/year.
Site microclimate effect:

Shaded location: microclimate effect of a shaded location is 5m/y which is added to the macroclimate
effect, from El Sarraf and Clifton (2011, section 4.3.1). Therefore, the design corrosion rate is now
25m/y.
Unwashed effects: taking into account the unwashed area Cuw multiplier from section 4.3.2, the multiplier
is:
Cuw = 1.2 for sites greater than 5km from the seacoast (option 5 of El Sarraf and Clifton 2011, section
4.3.2).
131
Therefore, the first-year corrosion rate is 25 x 1.2 = 30 m/y.

10.2.3.3 Corrosivity category for the site, including microclimate effects
This is obtained by using the first-year corrosion rate calculated in table 10.1 as 30 m/y.
From table 10.1, the corrosivity category C20%D applies; this designation means that the determined
atmospheric corrosivity category is within C and 20% of the way towards category D.
Table 10.1
Corrosivity categories
Corrosion rate for steel m/year
AS/NZS 2312
ISO 9223
<1.3
A: Very low
C1
1.3 to 25
B: Low
C2
25 to 50
C: Medium
C3
50 to 80
D: High
C4
80 to >200
E: Very high
C5
Note: This is based on table B1 of AS/NZS 2312 (SNZ 2002).
10.2.3.4 Determine the time to first (major) maintenance required for the coatings system
For a more sustainable durability design option a time to first maintenance of 35 years or longer time is
sought, while a time to first maintenance of 25 years is considered for the standard durability design.
10.2.4 Select an appropriate corrosion protection system

A single coat inorganic zinc silicate (IZS) solvent borne (SB), IZS3-SB which has 125m dry film thickness will
meet the 35 years requirement. Based on El Sarraf and Clifton (2011) and AS/NZS 2312, for the determined
corrosivity category of C20%D, the time to first maintenance for the IZS3-SB system is 37 years.
Note that inorganic zinc silicate (IZS) coatings are available in either solvent-borne (SB or water-borne (WB)
form. AS/NZS 2312 gives lower performance levels for the solvent-borne (SB coatings than for the waterborne (WB coatings. However, this is misleading for the SB systems for the reasons given in section 7.3.5
of El Sarraf and Clifton (2011). The sacrificial protection offered by both products is similar; hence the only
difference in design life is due to the slightly increased quantity of active ingredient (powdered metallic
zinc) in the WB systems.
For the 25 years time to first maintenance, a polyurethane (PUR3) three coat thickness with a total 250m
DFT is chosen.
10.2.5 Determination of life cycle cost of IZS3-SB and PUR3

This section gives the life cycle cost estimate for the IZS3-SB and PUR3 coating systems over the proposed
100-year life determined above. The information is presented in figures 3.2 and 3.3 (see El Sarraf and
Clifton 2011 for further guidance). The estimated time to next maintenance is determined in accordance
with section 10.2.4 above. The coating and labour rates are provided from the coating supplier,
International Protective Coatings. The labour costs are based on shop application of large areas, with
adjustments made for site conditions as specified in note 6 in figures 3.2 and 3.3. The net present value is
then calculated to provide a more realistic associated cost.

10.2.5.1 Estimated time to next maintenance
The estimated time to next maintenance is assessed in accordance with AS/NZS 2312 and (Reina et al
1998) as follows:
132
10
After each painting (initial coat and maintenance recoats), 2% of the area is assumed to require touch up
after three years and another 5% of the area is assumed to require touch up at 75% of the time for which
the next full coating is required.
The time to first full coating is taken as 35 years in figure 3.2 and 25 years in figure 3.3. The time to
subsequent touch-up repair is taken as 75% of 35 years (26 years) and 75% of 25 years (18 years).
The two touch-up coats between full recoats are in accordance with the recommendations for bridge
coatings, based on experience. In practice, one or both may not be necessary. It is assumed that both
coatings thickness loss will be reasonably consistent over the exposed surface area so that all areas will
require reinstatement when the durability time to first maintenance is reached. For that reason, a full
recoat is specified when that time is reached. In practice, using the criteria for assessing when to paint or
repair from clause 10.2(a) of AS/NZS 2312, a longer interval between recoats than that used in figures 3.2
and 3.4 might be obtained.
133
Figure 10.2
Life cycle cost estimate for IZS3-SB including inspection
Operation Number
Year (1)
System Designation
%Area Maintained
0
1
2
3
4
5
6
7
8
9
10
11
0
0
3
26
26
35
35
38
61
61
70
70
A1+A2+F
B+F
C+F
E
C+F
E
D+F
C+F
E
C+F
E
D+F
100%
2%
2%
100%
5%
100%
100%
2%
100%
5%
100%
100%
Current Cost (2)

$/total m2
44.00
1.78
1.78
covered
4.46
covered
57.00
1.78
covered
4.46
covered
57.00
Total $
NPV (3) for DR (4)

8% $/total m2
44.00
1.78
1.42
client
0.60
client
3.86
0.10
client
0.04
client
0.26
52.06
Coating System Specification:

Based on 125 micron IZS3-SB, paint reference number CO1a from (AS/NZS2312 2002).
Time to first maintenance (35, 40, 50 years)
35
System Designation:
A1 Washdown
High pressure washdown to remove fabrication oil etc. Prior to blast cleaning
A2 Initial Coat:
Shop: Grit Blast to Class 2.5; 125 micron nominal DFT IZS3-SB.
B 2% Erection Touch-up:
Field: Abrasive blast of damaged area to Class 2.5; 125 micron nominal DFT IZS3-SB
C 2% or 5% Touch-up Repair:
Field: Abrasive blast of degraded area to Class 2.5; 125 micron nominal DFT IZS3-SB
D 100% Recoat:
Field: Abrasive blast of whole area to Class 2.5; 125-150 micron nominal DFT IZS3-SB
E Inspection of surface to determine condition 9: Field: Independent inspection of existing surface to determine extent of maintenance required at 75%, 100% tfm.
Cost covered by Transit NZ/ONTRACK general bridge inspection regime and not part of specific steel coating cost
F Independent inspection of coating 10:
Field: Independent inspection of applied coating for designations A2, B, C and D
Operation
Initial washdown in shop
Blast clean and apply single coat (labour cost)
Shop inspection of surface or coating
Site inspection of surface or coating
Base costings used are as follows

(these are for shop application of
large areas, with small area and site access
factors included for the on site work
as described below)
Cost $/m2 steel surface

4.00
18.00
4.00
5.00
Paint/Inspection
System Designation
Base Rate Paint $/m2
Base Rate Labour

$/m2
Area Factor Labour(5) $/m2
Initial Rate(6)
$/m2
Site Access Rate(7)

$/m2
Current Cost(8)
$/m2
A1
A2+F
B+F
C+F
0.00
18.00
19.00
19.00
4.00
22.00
23.00
23.00
1.00
1.00
2.40
2.40
4.00
40.00
74.20
74.20
0.00
0.00
15.00
15.00
4.00
40.00
89.20
89.20
D+F
E
19.00
0.00
23.00
5.00
1.00
1.00
42.00
5.00
15.00
15.00
57.00
20.00
Summary:
Initial Cost/total m2
NPV Maintenance Cost 100 year/total m2
Full life cycle NPV cost 100 year/total m2 $
$ 47.20
4.86
52.06
This covers operations 0, 1 and 2 above which are part of the construction contract
This covers operations 3 to 11
Notes:
1. Year after commissioning bridge.
2. Current Cost/total m2 (% Area Maintained x System Cost $/m2)
3. NPV Net Present Value: NPV = Cost/(1+DR/100)Y ear
4. Recommended discount rate for NZTA NZ is 8% for 30 years. (Economic Evaluation Manual (Volume 1) 2010)
5. Area factor for labour is obtained from Section 10.5 of (El Sarraf and Clifton 2011). It allows for higher setup and wastage costs due to the small areas involved.
6. The initial rate is the materials + labour including area factor.
7. The Site access rate is taken from Section 10.5 of (El Sarraf and Clifton 2011), using moderate for touch-up, small area maintenance and for recoat. This reflect
the remotness of the site.
8. The current cost/m2 is the initial rate plus the site access rate.
10. Inspection costs for the painted surface cover independent inspection of the area that has been painted.
DISCLAIMER: The above costings are approximate and should be confirmed with a coating supplier and applicator.
Source: El Sarraf and Clifton (2011).
134
10
Figure 10.3
Life cycle cost estimate for PUR3 including inspection
Operation Number
Year (1)
System Designation
%Area Maintained
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
0
0
3
18
18
25
25
28
43
43
50
50
53
68
68
75
75
A1+A2+F
B+F
C+F
E
C+F
E
D+F
C+F
E
C+F
E
D+F
C+F
E
C+F
E
D+F
100%
2%
2%
100%
5%
100%
100%
2%
100%
5%
100%
100%
2%
100%
5%
100%
100%
Current Cost (2)

$/total m2
60.50
2.52
2.52
covered
6.30
covered
73.50
2.52
covered
6.30
covered
73.50
2.52
covered
6.30
covered
73.50
Total $
NPV (3) for DR (4)

8% $/total m2
60.50
2.52
2.00
client
1.58
client
10.73
client
client
0.23
client
1.57
0.04
client
0.03
client
0.23
79.43
Coating System Specification:

Based on a 3 coat total DFT of 250 micron PUR3, paint reference number C06+C13+C26 from (AS/NZS2312 202) .
Time to first maintenance (35, 40, 50 years)
25
System Designation:
A1 Washdown
High pressure washdown to remove fabrication oil etc. Prior to blast cleaning
A2 Initial Coat:
Shop: Grit Blast to Class 2.5; 250 micron nominal DFT PUR3
B 2% Erection Touch-up:
Field: Abrasive blast of damaged area to Class 2.5; 250 micron nominal DFT PUR3
C 2% or 5% Touch-up Repair:
Field: Abrasive blast of degraded area to Class 2.5; 250 micron nominal DFT PUR3
D 100% Recoat:
Field: Abrasive blast of whole area to Class 2.5; 250 micron nominal DFT PUR3
E Inspection of surface to determine condition 9: Field: Independent inspection of existing surface to determine extent of maintenance required at 75%, 100% tfm
Cost covered by Client general bridge inspection regime and not part of specific steel coating cost
F Independent inspection of coating 10:
Field: Independent inspection of applied coating for designations A2, B, C and D
Base costings used are as follows

(these are for shop application of
large areas, with small area and site access
factors included for the on site work
as described below)
Operation
Initial washdown in shop
Blast clean and apply three coat system(labour cost)
Shop inspection of surface or coating
Site inspection of surface or coating
Cost $/m2 steel surface

4.00
32.50
4.00
5.00
Paint/Inspection
System Designation
Base Rate Paint $/m2
Base Rate Labour

$/m2
Area Factor Labour(5) $/m2
Initial Rate(6)
$/m2
Site Access Rate(7)

$/m2
Current Cost(8)
$/m2
A1
A2+F
B+F
C+F
D+F
E
0.00
20.00
21.00
21.00
21.00
0.00
4.00
36.50
37.50
37.50
37.50
5.00
1.00
1.00
2.40
2.40
1.00
1.00
4.00
56.50
111.00
111.00
58.50
5.00
0.00
0.00
15.00
15.00
15.00
15.00
4.00
56.50
126.00
126.00
73.50
20.00
Summary:
Initial Cost/total m2
NPV Maintenance Cost 100 year/total m2
Full life cycle NPV cost 100 year/total m2
65.02
14.41
79.43
This covers operations 0, 1 and 2 above which are part of the construction contract.
This covers operations 3 to 11.
Notes:
1. Year after commissioning bridge.
2. Current Cost/total m2 (% Area Maintained x System Cost $/m2)
3. NPV Net Present Value: NPV = Cost/(1+DR/100)Y ear
4. Recommended discount rate for NZTA NZ is 8% for 30 years. (Economic Evaluation Manual (Volume 1) 2010)
5. Area factor for labour is obtained from Section 10.5 of (El Sarraf and Clifton 2011). It allows for higher setup and wastage costs due to the small areas involved.
6. The initial rate is the materials + labour including area factor.
7. The Site access rate is taken from Section 10.5 of (El Sarraf and Clifton 2011), using moderate for touch-up, small area maintenance and for recoat. This reflect
the remotness of the site.
8. The current cost/m2 is the initial rate plus the site access rate.
10. Inspection costs for the painted surface cover independent inspection of the area that has been painted.
DISCLAIMER: The above costings are approximate and should be confirmed with a coating supplier and applicator.
Source: El Sarraf and Clifton (2011).
135
10.2.6 Determining the corrosion rate of weathering steel for a 100-year design
life
The calculation below will estimate the corrosion rate of the weathering steel based on El Sarraf and
Clifton (2011) and El Sarraf and Clifton (2005) and ISO 9224 (ISO 2012). It uses the corrosion rate of mild
steel determined in section 10.2.3 herein and that given in ISO 9224 to interpolate the corrosion rate of
weathering steel in comparison to mild steel. Therefore, the corrosion rate of weathering steel is:
10.2.6.1 Determine the corrosion rate of weathering steel for the first 10 years of the 100-year
design life from ISO9224 (ISO 1992)
From ISO 9224, the corrosion rate (rav) for the first 10 years for weathering steel for an atmospheric
corrosivity category (ACC) of C20%D is given as 2 to 8. Therefore, taking 20% into that range the first 10year corrosion rate is given as:
rav = 3.2 m/y

10.2.6.2 Determine the corrosion rate of weathering steel for the remaining 90 years of the 100- year
design life from ISO 9224
From ISO 9224, the corrosion rate (rlin) for the remaining 90 years for weathering steel for an ACC of
C20%D is given as 1 to 5. Therefore, taking 20% into that range the remaining 90-year corrosion rate is
given as:
r lin = 1.8 m/y

Therefore, the 100-year design life total corrosion per exposed face is:
3.2(10) + 1.8(90) = 194 m/exposed face or 0.2mm/exposed face
A factor of 2 is recommended to be applied to allow for localised increased rates of corrosion. Therefore,
when designing the main girders, a loss of 0.4mm/exposed face should be taken into account when
determining the section capacity of the girder.
10.2.6.3 Notes on using weathering steel
The calculated corrosion rate stated above is based on the protective patina layer forming, if this does not
form properly then continuous corrosion of the steel will occur which will be higher than the estimated
corrosion rate above. Weathering steel will start out as a rusty red colour but with time it will become a
darker earthy tone (nearly black) but only if the conditions are favourable (eg not subject to salt
contamination and low times of wetness). For the bridge in this example, the proposed site is an ideal
location for the protective patina layer to form.
10.2.6.4 Costs of using weathering steel
On average the cost of weathering steel is $300/tonne more than that of mild steel. In this case, the cost
(as at October 2008) of uncoated mild steel plate is $2200/tonne, this equates to an average cost of
$2500/tonne for weathering steel (the designer should contact the steel supplier for the latest prices of
mild and weathering steel when undertaking a cost comparison).
In this example, the cost per square metre of using weathering steel equates to $14.27/m2 to $31.2/m2
for beams ranging from an 800WB122 to 1200WB455 respectively.

Once the patina forms on the weathering steel, the greatly reduced corrosion rate will allow the steel to
meet the required performance requirements with negligible, if any, maintenance. These figures are
applicable for the net present value minus the inspection cost, as they equate to the initial cost of a
136
10
protective corrosion system. Even though weathering steel has this cost premium in comparison to
uncoated mild steel, the future cost savings on maintenance makes this option cost competitive.
10.2.7 Results discussion

A summary of the results is given in table 10.2. By comparing both coating systems net present value, the
use of a single coat coating with a longer time to first maintenance provides the most cost-effective
solution. This solution also has less on-site wastage with the fact that only one coating layer is needed to
be reinstated while the multi-coat option requires up to three coats. Also, with a longer time to first
maintenance the IZS3 option requires only two recoats in comparison with the three recoats for the PUR3,
this equates to a reduced possibility of accidents and other health and safety issues occurring.
However, when comparing the costs between the inorganic zinc silicates (IZS) single-coat system with the
weathering steel option, the results show that the latter provides a more economical solution. Also, with
the negligible maintenance, on-site wastage is further reduced to a minimum, while the health and safety
benefits have been greatly improved due to that negligible maintenance.
Table 10.2
Cost difference between the different corrosion protection systems
Corrosion
protection system
Cost ($/m2)* for a 100-year design life

Net present value
Notes
IZS3
$52.06
Single-coat system
PUR3
$79.43
Multi-coat system
Weathering steel
$31.2
Assuming a 1200WB 455
Note: Cost comparison is based on October 2008 prices.
10.2.8 Durability design example conclusion

Even though weathering steel produced the most economical solution, other factors must be considered
by the designer when specifying the optimum sustainable durability option. The designer must not only
consider achieving the most economic time to first maintenance option, but must also consider the
required structures performance, aesthetics, design life and location, which includes future maintenance
accessibility. These factors may govern the design and a more expensive option may be chosen to meet
requirements. Weathering steel starts as a layered rust colour which gets darker with age. Aesthetically
this may not be desirable in high-profile projects. IZS provides a single colour option of grey, while the
polyurethane option provides a wide range of colours and graffiti protection. The aesthetic factor is one of
many factors considered when carrying out a sustainable durability design. All these factors govern the
chosen corrosion protection system based on the requirement stated by the client. Whether it is a major
highway bridge project, a road bridge or a back country overpass, each structure has its challenges and
requirements which the designer must consider to produce the optimum solution for that structure.
10.3 Coating application and inspection

The correct application of a specified corrosion protection coating system is one of the crucial steps in the
protection of structural steelwork. It is recommended that the protective coatings are applied by a
qualified applicator who has acquired the required level of training on the application of coating systems
by a recognised organisation, such as Extractive Industries Training Organisation. Furthermore, regular
inspection throughout the application process by a certified independent third party inspector, either by a
Certified Board for Inspection Personnel, Australasian Corrosion Association or NACE International coating
137
inspector, is recommended to ensure the appropriate quality control and health and safety procedures are
followed. Most importantly, inspectors will also supervise the actual application of the coatings to approve
the correct coating with its resultant dry film thickness has been applied in accordance with the relevant
standards and coating suppliers guidelines.
Guidance on the specification, application and inspection of coating systems is available through Steel
Bridge Development Group on www.steelbridges.org.nz. Designers should be familiar with these
documents to ensure that the specified protective coating system will provide the required corrosion
protection, which includes its maintenance, throughout the design life of the structure.
10.4 Maintenance management

Another important topic is the maintenance management of coating systems and structures. All structures
require regular maintenance to guarantee their performance over the design life of the structure, as
specified in the Building code (DBH 1992) and the Bridge manual. However, this is sometimes not
adequately addressed at the design stage. Considerations for maintenance must be conducted as part of
the design process; this includes accessibility for future maintenance and the establishment of a
maintenance regime as required for all of the components during the life of the structure, from
reinstatement of the coating system to the regular cleaning of the steelwork of chemical and biogenic
contamination. This also applies to concrete elements of the composite bridge.
In reality, both steel and concrete bridges require periodic maintenance to attain a 100-year life. Where the
maintenance programmes will differ is when the work is required.
For example, concrete bridges in New Zealand have not been maintenance free. Significant remedial work
on many bridges, including high-profile bridges, has been required after 30 to 60 years with some cases
requiring a complete rebuild. Examples are the Newmarket Viaduct, for which the distortional effects of
temperature were under recognised by the design procedure at that time, resulting in the initial post
tensioning being insufficient to prevent tension on lower regions of the section near the supports.
Remedial work to suppress this, in the form of additional post tensioning, led to over-compression and
resulting deterioration of the concrete at the supports. Another bridge is the Victoria Park Bridge flyover,
which is suffering alkali-aggregate reaction and needs replacement and a number of bridges close to the
sea that have accelerated rebar corrosion due to improper concrete cover thickness (Bruce et al 1999;
Bruce et al 2006; Rogers et al 2009).
Steel bridges in New Zealand have also been adversely affected by shortcomings in the design procedures
of the day. The best-known example is the Auckland Harbour Bridge, which suffers from problems with
fluctuating loads in the deck of the clip-on box girders due to the effect of fatigue on the light-weight steel
deck not being adequately accounted for in the design procedures of the 1960s. However, steel bridges
can generally be repaired rather than needing to be replaced in this situation. In most cases, regular
inspection of the structure can determine potential structural issues more easily than concrete structures,
which allows for remedial actions to be undertaken before the issues become irreparable.
138
11
Other types of composite construction
11
11.1 Composite box girders

Steel-concrete composite box girders offer an attractive and economic form of construction for medium
span bridges ranging between 50m and 100m. They are commonly used for spans where plate girder
sizes may be excessive or where torsion, curvature or wind and seismic forces demand greater torsional
stiffness. Torsional stiffness is provided by the hollow box shape of the girder that is either rectangular or
trapezoidal in cross section. Basically, the box girder comprises two webs that are connected by a single
bottom flange, while each web has its own top flange for an open box girder or another single top flange
for a closed box girder.
Stiffeners are evenly spaced along the length of the girder and usually a single wide box girder is
sufficient for a single or double lane bridge. For wider bridges, two or more box girders can be used, with
additional restraints provided, if required at supports, between them.
The benefits of a steel-concrete composite girder are:
longer spans possible providing cost effect spans
span-long girder sections can be erected by mobile cranes or launched from one end of the bridge
enhanced torsional performance may reduce bearing requirements in comparison to a plate girder
bridge
smooth clean lines with clean surfaces provide an aesthetically pleasing structure
sloping webs and lack of bottom flange at the outer face reduced potential durability issues due to
water entrapment.
Disadvantages of box girder are:
difficult to fabricate thereby adding to the cost
more difficult to maintain due to the access to confined spaces inside the girder.
An example of a box girder is shown in figure 11.1.

Figure 11.1
Steel-concrete composite box girder
139
11.2 Network arch bridges

A network arch is a tied arch structure where the hangers supporting the tie, which also acts as the deck
of the bridge, are inclined and arranged in such a way that they cross one another at least twice. Such
structures are used to carry a roadway, railway or footpath.
Like any tied arch, the load on the deck is carried principally as compression in the arch and tension in the
tie. Increasing the rise of the arch reduces the axial forces in both the arch and the tie. The majority of the
shear force is taken by the vertical component of the arch top chord force, with any variation in the shear
force taken by the hangers.
This form of structure was developed in Norway during the 1950s by Per Tviet with the first network arch
bridge constructed at Steinkjer, Norway in 1963 (Tviet 2008). Although many examples of network arch
bridges have been constructed in Japan, where they are often (incorrectly) referred to as Neilsen-Lhose
bridges, network arch bridges have not been adopted widely around the world until recently. This is being
changed now with a number of network arches in design and construction in recent years across Europe,
the USA and now in New Zealand with the Mangamuhu Bridge on the Mangawhero River north-east of
Wanganui, and the Waikato River Bridge near Taupo (figure 11.3).
Figure 11.2
Waikato Network Arch Bridge near Taupo (image courtesy of Holmes Consulting Group)
Network arch bridges tend to act like a truss with a light web. With inclined hangers, the effects of
concentrated or non-symmetric loading are distributed better than with vertical hangers and the bending
moments and shear forces in the arch and deck due to such loading are reduced.
The benefits of using an inclined overlapping hanger system may be seen in figures 11.3 and 11.4. When the
span is subjected to a distributed load on only part of the span length, some of the hangers will tend to relax
as can be seen in figure 11.3. Increasing the distance: between hanger nodes reduces the tendency for the
hangers to relax but this creates increased bending in the arch chords requiring larger chord elements.
Figure 11.3
Partial loading causing relaxation of hangers (Tviet 2008)
Relaxed hangers
140
11
As the arch top chord tends to deflect upward and the bottom chord downwards between the arch hanger
points under loading, introducing a second set of inclined hangers between the first set helps to minimise
the arch chord deflections and bending actions as seen in figure 11.4. Several sets of overlapping hangers
can be introduced to decrease the distance between hanger points further, thus creating the network arch
configuration.
Figure 11.4
Additional hangers added to original layout (Tviet 2008)
Additional set of hangers
By decreasing the distance between hanger locations, bending actions in the archs top and bottom chord
elements are reduced. The shorter length of the arch top chord also increases the buckling strength of the
top chord element. By minimising the bending component of action in the arch chords, greater efficiency
can be realised from a given chord element section. Compared with similar span arches using vertical
hangers, reported savings of up to 50% of the structural steel weight can be realised through a network
arch structure (Tviet 2008).
The arch top chord can be profiled to follow the line of thrust of forces within the arch. For simplicity of
fabrication, it is easy to adopt a circular geometry for the arch top chord profile. While compressive
actions dominate in the arch top chord, good support is offered by the hangers reducing the tendency of
the chord to buckle in the vertical plane. Therefore greater stiffness is provided in the transverse direction,
leading to the adoption of UC or H-profile sections for these elements in a number of bridges, where heavy
rolled UC or H-profile sections are available, such as in Europe and the USA, these are very efficient
choices for top chord elements. As heavy hot-rolled UC sections (upwards of 400 600kg/m) are not
generally available in New Zealand, some fabrication of the top chord may be required.
The original concept prefers that the bottom tie is constructed from post-tensioned concrete the tensile
force carried in longitudinal post-tensioning tendons. The compression provided to the concrete slab
limits cracking of the bottom tie member. Depending on the width of the bridge between arches, the
bottom concrete deck tie may be transversely post-tensioned as well.
Depending on construction methodology, some form of temporary bottom tie may be required (for
example, if the steelwork tie is erected and lifted into place prior to the bottom concrete tie being
completed). For Mangamuhu Bridge, it was found that it was appropriate to utilise a ladder form of deck
construction with a steel UC longitudinal tie and transverse transoms at regular centres (Chan and
Romanes 2008). The precast deck slabs were made composite with the transoms and bottom tie with an in
situ topping. Longitudinal post-tensioning is provided within the deck slab to provide a residual
compressive stress and control the variable tensile stresses due to live loading.
11.3 Stainless steel in bridge construction

Stainless steel is becoming a more common choice for bridge construction around the world and
especially in Europe, as engineers begin to recognise the benefits that the materials durability can offer. A
number of bridges incorporating stainless steel for the main structural elements have been built in the
141
past eight years, such as the recently opened Aorga Railway Bridge in Spain (figure 11.5), Likholefossen
Bridge in Norway (figure 11.6), the Celtic Gateway Bridge in Holyhead, Wales (figure 11.7), and the
Passarella Ruffolo in Sienna, Italy.
Figure 11.5
Aorga Railway Bridge in Spain (photo courtesy of Outokumpu)
Figure 11.6
Likholefossen Bridge in Norway
Figure 11.7
The Celtic Gateway Bridge in Holyhead, Wales
142
11
Although there is an enormous variety in stainless steel types, their common denominator is the presence
of at least 11% of chromium. In combination with other elements such as nickel, molybdenum or nitrogen,
this produces a steel alloy that has high corrosion resistance, malleability, ductility and mechanical
strength, even when exposed to high temperatures, as well as excellent aesthetics and easy maintenance
and cleaning. The chromium in the stainless steel forms a stable and transparent layer of chromium oxide
on the surface - the so-called passivation layer - that prevents corrosion. Four main types of stainless
steel exist according to their metallurgical structure: ferritic, austenitic, duplex and martensitic - of these,
the high strength of duplex steel makes it suitable for applications such as bridges.
Duplex stainless steel is an austenitic-ferritic alloy which has a microstructure of high corrosion resistance,
excellent ductility and mechanical characteristics superior to the great majority of carbon steels. With the
existence of such a wide range of duplex steel grades, the selection of the most suitable type clearly
depends on the ambient aggressiveness, mechanical properties, types of surface finish and so forth.
Unlike conventional carbon steel, stainless steel exhibits a nonlinear mechanical behaviour, even under
lower stress values, without having a clearly defined elastic limit strength. However, it is being designed
with a conventional yield stress with the value associated to a strain of 0.2%. The construction procedures
for stainless steel are similar, but not identical to, those used for carbon steel. Austenitic steel shows
excellent possibilities for bending, but it requires 50% more energy than carbon steel. It has a similar
energy requirement when being welded, making it difficult to weld duplex steel grades. In addition,
contact between stainless steel and other metals during manufacture can cause iron particles to be
embedded in the stainless steel surface, thereby penetrating the protective passivation layer and initiating
corrosion. Galvanic corrosion can also occur due to iron contamination at any stage, for example, the use
of carbon steel strapping on stainless steel components or the bolting of carbon steel onto stainless steel.
For this reason, manufacturing and assembly of the pieces must be carried out in areas that will not come
in contact with carbon steel, and has to be done using specific tools.
In 2005, stainless steel was chosen for the Cala Galdana Bridge on the island of Menorca - the location of
the bridge, in a harsh and corrosive environment, made stainless steel, with its resistance to corrosion, an
attractive choice. The Algendar river runs into the sea at Cala Galdana and its channel has for the last 30
years been crossed via a reinforced concrete bridge approximately 18m long, but the concrete bridge
reinforcement bars were in an advanced state of corrosion, induced by the marine atmosphere and serious
settlement of one of the abutments. This promoted the owner Consell Insular de Menorca to replace it
with a new bridge. The new bridge had to span the entire width of the old river channel, more than 40m,
fit in with the natural surroundings, demonstrate great durability and require minimum maintenance. The
final choice of a duplex stainless steel arch structure was mainly due to its high resistance to corrosion. In
addition, its clear span restored the full width of the river channel, offered minimum maintenance and was
regarded as a symbol of modern technology.
The overall length of the bridge is 55m with a 13m wide deck, two lanes of road traffic and two footways
each 2m wide, to allow pedestrians to enjoy the panoramic views. The main structure consists of two
parallel abutments by means of an inclined strut that takes the horizontal component of the arch axial
force and, consequently, does not transmit significant horizontal forces to the abutments. The weight of
the stainless steel is approximately 165 tonnes and the total cost of the bridge, including its accesses, was
approximately 2.6 million.
143
Figure 11.8
The underside of the Cala Galdana Bridge
11.3.1 Are stainless steel bridges an option for New Zealand?

The answer is yes. The excellent durability performance and aesthetic properties of stainless steel lends
its use for certain areas around New Zealand, specifically the areas within 5km from the sea. Initial
construction cost is relatively higher than other materials; whether it is mild steel, weathering steel or
concrete; however, by taking the life cycle cost of a bridge including the maintenance cost over the 100year design life, stainless steel can provide a competitive solution in those specific locations. Stainless
steel does lend itself to iconic projects and especially for pedestrian bridges. Lean duplex grade LDX 2101
is a cost-effective option in comparison with normal duplex stainless steel, for pedestrian bridges, while
Duplex Grade 2205 is preferred for road or rail bridges.
As with other materials, the design must specify the correct specification which is crucial for the success
of the project. One of the concerns with stainless steel is the discolouration of the surface of the steel
known as tea staining. This is caused by the use of a rougher surface finish than that required, especially
in areas close to the sea. To take maximum advantage of the benefits of stainless steel, the correct grade
and surface finish must be specified depending on the location and use of the bridge structure.
Guidance on the structural design of stainless steel is available via New Zealand Heavy Engineering
Research Association (HERA), the New Zealand Stainless Steel Development Association and excellent
guidance is available from the American Nickel Institute and the British Stainless Construction websites.
Also guidance on the welding of structural stainless steel is given in AS/NZS 1554.6 (SNZ 1994).
144
12
References
12
References
American Iron and Steel Institute (AISI) (2001) Designing short span bridge. Washington DC: AISI.
Australian Steel Institute (ASI) (1999) Design capacity tables for structural steel. Sydney: Australian Steel
Institute.
Beamish, M, P Tindall and P Billing (2006) Auckland Harbour Bridge fatigue assessments. In Proceedings of
2006 Austroads Conference. Perth, Australia, 2006.

Beer, FP, R Johnston Jr, J DeWolf and D Mazurek (2008) Mechanics of materials. McGraw-Hill. New York.
Bird, G and M Feeney (1999) Simplified design of steel members. Auckland: SESOC.
British Constructional Steelwork Association (BCSA) (2005) BCSA guide to the erection of steel bridges. (ref
38/05).
Bruce, SM, SA Freitag and WE Hickman (1999) Durability of concrete road bridges in New Zealand.
Transfund NZ research report 129. 87pp.

Bruce, SM, PS McCarten, SA Freitag and LM Hasson (2006) Deterioration of prestressed concrete bridge
beams. Land Transport NZ research report 337.
Cement and Concrete Association of New Zealand (CCANZ) Concrete3 economic, social and environmental.
Wellington: CCANZ.
Chan, M and M Romanes (2008) New Zealands first network arch bridges. 2008 Metals Industry
Conference Proceedings, HERA, Manukau City, New Zealand.

Clifton, GC (1997) Restraint classifications for beam member moment capacity determination to NZS
3404:1997. HERA report R4-92.
Clifton, GC (2007a) Recommended draft fatigue design criteria for bridges. Manukau City: HERA.
Clifton, GC (2007b) New Zealand structural steelwork limit state design guides volume 1. HERA report R4
80.
Corus (2006) Sustainable steel construction. Scunthorpe, UK: Corus.
CSC World (2006) S-Frame software. Leeds, UK: CSC.
Department for Transport (2006) TD19/06: Requirement for road restraint systems. Design manual for
roads and bridges. Accessed 27 April 2013.

www.dft.gov.uk/ha/standards/dmrb/vol2/section2/td1906.pdf
Department of Building and Housing (DBH) (1992) New Zealand building code. Wellington: Building
Industry Authority, DBH.
Duncan, J and R Cordner (1991) Atmospheric corrosion rates over two years' exposure at 156 sites in New
Zealand. IPENZ transactions 18 (1/GEN), 1991, pp3749.
El Sarraf, R and GC Clifton (2005) New Zealand weathering steel guide for bridges. HERA report R497.
El Sarraf, R (2008) Economical steel bridge solutions for New Zealand. SESOC Journal 21, no.2.
El Sarraf, R and GC Clifton (2011) New Zealand steelwork corrosion and coatings guide. HERA report R4
133.
Fisher, J (1977) Bridge fatigue guide: design and details. Chicago: American Institute of Steel Construction.
145
Gorenc, B, R Tinyou and A Syam (2005) Steel designers handbook. Sydney: UNSW Press.
Hambly, EC (1990) Bridge deck behaviour. Spon Press. UK.
Hayward, A (2002) Composite steel highway bridges. Scunthorpe, UK: Corus Construction Centre.
Heath and Safety Executive (2007) Managing health and safety in construction. Construction (Design and
Management) Regulations. L144. UK: HSE Books.
Hendy, CR and DC Iles (Eds) (2009) Steel bridge group: guidance notes on best practice in steel bridge
construction. SCI P185 (5th issue). Ascot, UK: The Steel Construction Institute.
HERA (2006); BRIDGENZ. Manukau City. Now withdrawn.
Hyland, CWK and M Enzensberger (1998) Prediction of site-specific steel corrosion rates in New Zealand to
assist coatings selection. In Proceedings of Australasian Structural Engineering Conference, volume 2,
pp835842, Auckland, New Zealand.
Iles, DC (2010) Composite highway bridge design. SCI P356. Ascot, UK: The Steel Construction Institute.
International Federation for Structural Concrete (fib) (2005) Guidelines for the design of footbridges.
Lausanne: fib.
Medland, IC and CM Segedin (1979) Brace forces in interbraced column structures. Journal of the
Structural Division 105, no.7: 15431556.

Ministry of Works (1981) Standard bridge design. Wellington: Ministry of Works.
Mutton, BR and NS Trahair (1975) Design requirements for column braces. CE17, no.1. Australia: Civil
Engineering Transactions, Institution of Engineers, pp3035.
National Roads Board (1984) Road Research Unit bulletin 70 (RRU 70). Wellington: National Roads Board.
New Zealand Steel (2003) MemDes version.2.1. Auckland: New Zealand Steel.
New Zealand Transport Agency (NZTA) (2001) Bridge inspection and maintenance manual, SP/M/016 (rev
ed). Wellington: NZTA.
New Zealand Transport Agency (NZTA) (2003) Bridge manual, SP/M/022. Wellington: NZTA.
New Zealand Transport Agency (NZTA) (2004) Provisional amendment to section 5 of the Bridge manual.
Wellington: NZTA.
New Zealand Transport Agency (NZTA) (2006) Guide to heavy vehicle management. RTS16. Wellington:
NZTA.
New Zealand Transport Agency (NZTA) (2011) Economic Evaluation manual (Volume 1). Wellington: NZTA.
New Zealand Transport Agency (NZTA) (2012) Bridge manual. 3rd ed. Wellington: NZTA.
Pucher, A (1964) Influence surfaces of elastic plates. Vienna: Springer-Verlag.
Rapattoni, F (1998) Composite steel road bridges concepts and design charts. Port Kembla, Australia: BHP.
Reina, MP, M Melampy and KR Shields (1998) Costing considerations for maintenance and new
construction coating work. Materials Performance, November 1998: 2530.
Rogers, M, MA Al-Ani and JM Ingham (2009) Prevalence and distribution of pre-tensioned prestressing
steel corrosion in the New Zealand concrete bridge stock. The New Zealand Concrete Industry
Conference, Rotorua, 810 October 2009.
146
12
References
Steel Construction New Zealand (SCNZ) (2007) Steel connect, SCNZ 14.1. Manukau City, New Zealand:
SCNZ.
Tata Steel International (2005) ComFlor 80. Accessed 8 June 2013. www.tatasteelnz.com/ComFlor_80.asp
Tviet, P (2008) The network arch bits of manuscript in September 2008 after lectures in 50 countries.
Accessed 27 April 2013. http://home.uia.no/pert/backup/documents/The_Network_Arch__Fall_2008.pdf
Vianna, J, L Costa-Neves, P Vellasco and S Andrade (2008) An experimental evaluation of composite
structure perforated shear connectors; Steel Bridges Advanced Solutions and Technologies. 7th
International Conference on Steel bridges. ECCS. Europe. II-125-II-132
Yandzio, E and DC Iles (2004) Precast concrete decks for composite highway bridges, SCI P316. Ascot, UK:
The Steel Construction Institute.
Standards cited in this report

British Standards Institution (BSI) (1980) BS 5400-10 Steel, concrete and composite bridges. Code of
practice for fatigue. London: BSI.

practice for design of steel bridges. London: BSI.

practice for design of composite bridges. London: BSI.

British Standards Institution (BSI) (2006) BS 5400-2 Steel, concrete and composite bridges part 2:
specification for loads. London: BSI.

British Standards Institution (BSI) (2008) PD 6695-2 Recommendations for the design of bridges to BS EN
1993. London: BSI.

British Standards Institution (BSI) (2011) PD 6694-1 Recommendations for the design of structure subject
to traffic loading to BS En 1997-1. London: BSI.

European Committee for Standardization (CEN) (2003) BS EN 19912 Eurocode 1: Actions on structures.
Brussels: CEN.
European Committee for Standardization (CEN) (2005a) EN 1993-19 Eurocode 3: Design of steel
structures part 19: fatigue. Brussels: CEN.

European Committee for Standardization (CEN) (2005b) EN 1994-2 Eurocode 4: Design of composite steel
and concrete structures part 2: general rules and rules for bridges. Brussels: CEN.
European Committee for Standardization (CEN) (2006) EN 1993-2 Eurocode 3: Design of steel structures
part 2: steel bridges. Brussels: CEN.

Highways Agency (2001) BD 49/01 Design rules for aerodynamic effects on bridges. London: Highways
Agency, Department for Transport. Accessed 27 April 2013.

www.dft.gov.uk/ha/standards/dmrb/vol1/section3/bd4901.pdf
Highways Agency (2006) TD 19/06 Requirement for road restraint systems. London: Highways Agency,
Department for Transport. Accessed 27 April 2013.
www.dft.gov.uk/ha/standards/dmrb/vol2/section2/td1906.pdf
147
Highways Agency (2010) BD 37/01 Loads for highway bridges. London: Highways Agency, Department for
Transport. Accessed 27 April 2013. www.standardsforhighways.co.uk/dmrb/vol1/section3/bd3701.pdf
International Organization for Standardization (ISO) (2012) ISO 9224 Corrosion of metals and alloys
corrosivity of atmospheres Guiding values for the corrosivity categories. Geneva: ISO.
Standards Australia (SA) (1998) AS 3828 Guidelines for the erection of building steelwork. Sydney:
Standards Australia.
Standards Australia (SA) (2004) AS 5100 Bridge standard. Including amendment no.1 (2010). Sydney:
Standards Australia. .
Standards New Zealand (SNZ) (1994) AS/NZS 1554.6 Structural steel welding welding stainless steels for
structural purposes. Wellington: SNZ.

Standards New Zealand (SNZ) (1996) AS/NZS 1252 High-strength steel bolts with associated nuts and
washers for structural engineering. Wellington: SNZ.

Standards New Zealand (SNZ) (1997a) NZS 3109 Concrete construction. Wellington: SNZ.
Standards New Zealand (SNZ) (1997b) NZS 3404 Steel structure standard, parts 1 and 2. Wellington: SNZ.
Standards New Zealand (SNZ) (2002) AS/NZS 2312 Guide to the protection of structural steel against
atmospheric corrosion by the use of protective coatings, incorporating amendment no.1:2004.

Wellington: SNZ.
Standards New Zealand (SNZ) (2003a) AS/NZS 1170.3 Structural design actions snow and ice actions.
Wellington: SNZ.
Standards New Zealand (SNZ) (2003b) NZS 3104 Specification for concrete production. Wellington: SNZ.
Standards New Zealand (SNZ) (2004a) AS/NZS 1170.5 Earthquake actions New Zealand. Wellington: SNZ.
Standards New Zealand (SNZ) (2004b) AS/NZS 1554 Structural steel welding set. Wellington: SNZ.
Standards New Zealand (SNZ) (2004c) AS/NZS 1554.1 Structural steel welding welding of steel structures.
Wellington: SNZ.
Standards New Zealand (SNZ) (2004d) AS/NZS 1554.5 Structural steel welding welding of steel structures
subject to high levels of loading. Wellington: SNZ.

Standards New Zealand (SNZ) (2004e) NZS 1170.5 Structural design actions earthquake actions New
Zealand. Wellington: SNZ.

Standards New Zealand (SNZ) (2006) NZS 3101 Concrete structures standard. Wellington: SNZ.
Standards New Zealand (SNZ) (2009) NZS 3404 Steel structure standard, part 1. Wellington: SNZ.
Standards New Zealand (SNZ) (2010a) AS/NZS 3679.1 Structural steel - hot-rolled bars and sections.
Wellington: SNZ.
Standards New Zealand (SNZ) (2010b) AS/NZS 3679.2 Structural steel welded I sections. Wellington: SNZ.
148
Appendix A: Two-lane, three and four girder bridge costs
Appendix A: Two-lane, three and four girder

bridge costs
A1
How to achieve a cost-effective steelconcrete composite bridge
While the following guidance is for short span bridges, the design philosophy is applicable to any span
length. This section starts by providing a summary of current international preliminary design methods
and the current design and construction practice in New Zealand, followed by a study on the cost
breakdown of a steel-concrete composite bridge. A cost-effective superstructure configuration is
determined and a cost comparison is given in an example from two real bridge projects where a steelconcrete composite bridge option was found to be more cost effective than a conventional concrete
bridge. Costs are given for the superstructure and substructure as well as the coating maintenance.
A2
Current short span design practice
The American Designing short bridges (AISI 2001) provides complete guidance on the number of girders,
girder section size, spacing and other required data for spans from 6m to 36m and decking width from
7m to 13m in a table format. While, the British Tata Steel publication (Hayward 2002) provides the
guidance in a chart format for spans up to 60m and decking width up to 16m.
Previous New Zealand guidance Standard bridge design (Ministry of Works 1981) followed the American
format; however, it specifies structural forms and solutions which are either now not available or costeffective and it is currently out of print, while Australias (Rapattoni et al 1998) follows the British format.
The main difference in scope and format is that the American solution provides a complete design
solution while the British charts provide preliminary sizing and require detailed design. The common
aspect of both these publications is that the substructure and the different decking options are not
considered. Work undertaken by El Sarraf and Clifton (2008) has shown that these aspects have a major
effect on the cost (as shown in the next section) and should be considered early in the design stage.
A3
Cost breakdown through an example
A3.1
Design parameters
The following design parameters and assumptions were used for the design example. Costs are also included.
A3.1.1
Deck widths and different deckings
The deck width was determined in accordance with the Bridge manual therefore, for a two-lane bridge
with a medium volume of average annual daily traffic (AADT) of 4000 vehicles, the width is 9.7m. A single
lane 6.2m wide option was also researched but the results are not included in this guide due to space
limitations. The full research findings are found in El Sarraf and Clifton (2008).
Three types of decking types were considered, namely:
in situ concrete deck
full depth precast concrete deck
partial depth precast concrete deck.

149
During the analysis of the bridge, each deck type incorporated the appropriate construction live load, as
shown in table A.1. For the precast decking construction cases, two deck placement options were used,
with the corresponding construction live loads. The first was based on the units being installed by crane
and the second by mobile excavator with lifting arm, requiring the excavator to drive across the bridge to
place the units. For the shorter spans especially, this second case is relatively common and imposes
higher constructional live loading, resulting in the need for larger beam sizes.
A3.1.2
Span
A total of five spans were considered to determine the difference in the main girder dimensions. The
spans chosen were 12m, 15m, 21m, 25m, 27m and 30m, which were selected based on feedback from
designers and fabricators. These are the most common spans and come in increments of 3m, as this
matches the available lengths of plate and avoids over-sizing.
A4
Load cases
All the bridge options have the same imposed vertical loadings, which are presented in table A.1.
Table A.1
Imposed vertical loadings on bridge
Item
Dimensions
Steel girders
Dependent on the span
Deck
200mm thick
4.8kN/m2
Superimposed load
1.5kN/m2
Construction live load (in situ decking)
0.75kN/m2
Construction live load (full depth precast)
0.5kN/m2
Construction live load (partial depth precast)
Load
0.75kN/m2
Live load UDL
3.5kN/m2
HN point load
120kN/axle
HO point load
240kN/axle
Excavator point load
80kN/axle
Note: The excavator used was a Caterpillar M313D wheel excavator.
A5
Design procedure
Once the factored applied load on each beam was determined, calculation of moments and shears was
straightforward. The maximum moment involved the axle loads symmetrically placed about the midspan, while maximum shear involved the axle loads close to the supports. A number of computer
programs were used to assist in the design of the different bridge configurations.
For the three and four girder options, the distribution of deck dead and live loads into the supporting
girders had to be considered. This was determined taking into account the position of the load, the
longitudinal stiffness of the girders and the transverse stiffness of the deck. The computer analysis
program S-Frame (CSC 2006) was used to model the bridge deck and determine the load distribution
factors, which are dependent on the load case and the location of the live load point loads and uniformly
distributed loads. Once the distribution factors were determined, the loads and moments affecting each
girder were calculated and the girder section size was chosen using the New Zealand produced design
150
programs MemDes (New Zealand Steel 2003) for the construction loading and BRIDGENZ (HERA 2006) for
the composite loading case (which is now withdrawn).
The section sizes were chosen based on a number of conditions:
1
The strength requirements, both during the construction and when the girders were acting
compositely with the decking, were considered, with the larger size chosen for the costing stage.
As the number of braces decreases as girder section size increases, this resulted in different options
being considered.
If the ratio of section design capacity/design action was 1.03, that section was chosen
In general, a lighter but deeper beam was chosen over a shallower but heavier beam.
A6
Costing used
Table A.2 shows the costing used for the different components. Note that the substructure cost and other
related costs other than those outlined in table A.2 were not considered in this study. Costing was
obtained from the bridging industry, ie a precast concrete manufacturer, a steel fabricator, coatings
supplier and contractor.
Table A.2
Costing used (in August 2008)
Item
Cost
In situ concrete deck
$1800/m3
Full depth precast concrete deck
$1650/m3
Partial depth precast concrete deck
$1320/m3
Coating system (IZS3SB)
$29/m2
Fabrication cost
$4700/tonne
Delivery, handling and erection cost
$700/tonne
Excavator
$1000/day
Craneage
$4000/day
The in situ concrete decking cost comprised: installed in place concrete, reinforcing bar, formwork and
temporary support. The full depth precast cost comprised the installed in-place decking and stitching
installation (connecting the decking together) and, finally, the partial depth precast decking cost
comprised the installed in-place decking units together with the concrete and the rebar. Rebar quantities
were based on 0.3% each way each face, being the minimum required by the Bridge manual.
The fabrication cost included the steel plate, shop drawing, welding, cutting and other related fabrication
costs; it did not include the protection coating cost, which was calculated separately. The excavator (for
installing the precast decking for the 12m and 15m span bridges) and craneage cost assumed a 10m span
use per day. The fabrication and delivery, handling and erection cost were added together to keep the
number of variables in table A.3 to a minimum. The concrete decking and fabricated steelwork pricing was
determined in accordance with common industry practice, which uses the per cubic metre cost for
concrete decking and per tonne cost for fabricated steelwork.

An X-brace set consists of two unequal angle sections, four pairs of M20 bolts and a welded plate where
the angles meet at the X-point. The assumed coating, selected to give a time to first maintenance of at
least 35 years for a location of atmospheric corrosivity category C, was a 125m of factory applied
151
inorganic zinc silicate (IZS3SB), with cost of NZD$29/m2 (in August 2008). The concrete decking type and
X-brace sizes are shown in table A.3.
Table A.3
Concrete deck and X-brace numbering system
Number
Item
In situ concrete deck
Full depth precast concrete deck with excavator
2*
Full depth precast concrete deck without excavator
Partial depth precast concrete deck with excavator
3*
Partial depth precast concrete deck without excavator
125x75x8 UA
125x75x10 UA
125x75x12 UA
150x90x12 UA
150x100x10 UA
150x100x12 UA
150x90x10 UA
152
Table A.4
Span
Girders
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
21
21
21
21
21
21
21
21
21
21
21
21
27
27
27
27
27
27
27
27
27
27
27
27
30
30
30
30
30
30
30
30
30
30
30
30
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
3
3
3
3
3
3
4
4
4
4
4
4
3
3
3
3
3
3
4
4
4
4
4
4
3
3
3
3
3
3
4
4
4
4
4
4
Deck
Type
1
2
2*
3
3*
1
2
2*
3
3*
1
2
2*
3
3*
1
2
2*
3
3*
1
2
2*
3
3*
1
2
2*
3
3*
1
2
2*
3
3*
1
2
2*
3
3*
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Two lane, three and four girder bridge costs

Deck Cost
$41,904.0
$38,412.0
$38,412.0
$30,729.6
$30,729.6
$41,904.0
$38,412.0
$38,612.0
$30,729.6
$30,929.6
$41,904.0
$38,412.0
$38,612.0
$30,729.6
$30,929.6
$41,904.0
$38,412.0
$38,612.0
$30,729.6
$30,929.6
$52,380.0
$48,015.0
$48,215.0
$38,412.0
$38,612.0
$52,380.0
$48,015.0
$48,215.0
$38,412.0
$38,612.0
$52,380.0
$48,015.0
$48,215.0
$38,412.0
$38,612.0
$52,380.0
$48,015.0
$48,215.0
$38,412.0
$38,612.0
$73,332
$67,221
$53,777
$73,332
$67,221
$53,777
$73,332
$67,221
$53,777
$73,332
$67,221
$53,777
$94,284
$86,427
$69,142
$94,284
$86,427
$69,142
$94,284
$86,427
$69,142
$94,284
$86,427
$69,142
$104,760
$96,030
$76,824
$104,760
$96,030
$76,824
$104,760
$96,030
$76,824
$104,760
$96,030
$76,824
No of
Braces Set
8
8
8
8
8
2
2
2
2
2
12
12
12
12
12
3
3
3
3
3
8
8
8
8
8
2
2
2
2
2
12
12
12
12
12
3
3
3
3
3
8
8
8
6
6
6
12
12
12
9
9
9
8
8
8
6
6
6
12
12
12
9
9
9
8
8
8
6
6
6
12
12
12
9
9
9
Bracing
Size
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
1
1
1
1
1
1
7
7
7
7
7
7
1
1
1
1
1
1
2
2
2
2
2
2
4
4
4
4
4
4
Choosen Girder
Size
800 WB 146
800 WB 146
800 WB 146
800 WB 146
800 WB 146
800 WB 146
800 WB 168
800 WB 146
800 WB 168
800 WB 146
610 UB 113
610 UB 113
610 UB 113
610 UB 113
610 UB 113
610 UB 113
700 WB 115
610 UB 113
700 WB 115
610 UB 113
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
1000 WB 215
800 WB 146
800 WB 146
800 WB 146
800 WB 146
800 WB 146
800 WB 146
900 WB 175
800 WB 146
900 WB 175
800 WB 146
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 249
1200 WB 249
1200 WB 249
1200 WB 249
1200 WB 249
1200 WB 249
1200 WB 423
1200 WB 423
1200 WB 423
1200 WB 423
1200 WB 423
1200 WB 423
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 278
1200 WB 636
1200 WB 636
1200 WB 636
1200 WB 636
1200 WB 636
1200 WB 636
1200 WB 392
1200 WB 392
1200 WB 392
1200 WB 392
1200 WB 392
1200 WB 392
Coating Cost
$3,513
$3,513
$3,513
$3,513
$3,513
$2,977
$2,998
$2,977
$2,998
$2,977
$3,648
$3,648
$3,648
$3,648
$3,648
$3,108
$3,472
$3,108
$3,472
$3,108
$4,858
$4,858
$4,858
$4,858
$4,858
$4,315
$4,315
$4,315
$4,315
$4,315
$5,397
$5,397
$5,397
$5,397
$5,397
$4,847
$5,364
$4,847
$5,364
$4,847
$7,506
$7,506
$7,506
$7,323
$7,323
$7,323
$9,072
$9,072
$9,072
$8,879
$8,879
$8,879
$10,956
$10,956
$10,956
$10,773
$10,773
$10,773
$12,383
$12,383
$12,383
$12,191
$12,191
$12,191
$14,115
$14,115
$14,115
$13,953
$13,953
$13,953
$16,181
$15,944
$15,944
$15,730
$15,730
$15,730
Fab/erecction
cost
$32,391
$32,391
$32,391
$32,391
$32,391
$29,385
$33,662
$29,385
$33,662
$29,385
$33,324
$33,324
$33,324
$33,324
$33,324
$30,298
$30,824
$30,298
$30,824
$30,298
$56,300
$56,300
$56,300
$56,300
$56,300
$53,259
$53,259
$53,259
$53,259
$53,259
$51,416
$51,416
$51,416
$51,416
$51,416
$48,332
$57,740
$48,332
$57,740
$48,332
$99,512
$99,512
$99,512
$98,278
$98,278
$98,278
$117,261
$117,261
$117,261
$116,182
$116,182
$116,182
$191,044
$191,044
$191,044
$189,538
$189,538
$189,538
$166,444
$166,444
$166,444
$165,365
$165,365
$165,365
$313,462
$313,462
$313,462
$312,370
$312,370
$312,370
$256,090
$262,785
$262,785
$260,593
$260,593
$260,593
153
Decking
erection cost
$0
$4,800
$1,200
$4,800
$1,200
$0
$4,800
$1,200
$4,800
$1,200
$0
$4,800
$1,200
$4,800
$1,200
$0
$4,800
$1,200
$4,800
$1,200
$0
$6,000
$1,500
$6,000
$1,500
$0
$6,000
$1,500
$6,000
$1,500
$0
$6,000
$1,500
$6,000
$1,500
$0
$6,000
$1,500
$6,000
$1,500
$0
$8,400
$8,400
$0
$8,400
$8,400
$0
$8,400
$8,400
$0
$8,400
$8,400
$0
$10,800
$10,800
$0
$10,800
$10,800
$0
$10,800
$10,800
$0
$10,800
$10,800
$0
$12,000
$12,000
$0
$12,000
$12,000
$0
$12,000
$12,000
$0
$12,000
$12,000
Total cost without deck

Cheapest
(incl steelwork)
option no deck
$35,904.54
$35,904.54
$35,904.54
$35,904.54
$35,904.54
$32,361.38
$32,361.38
$36,659.67
$32,361.38
$36,659.67
$32,361.38
$36,971.40
$36,971.40
$36,971.40
$36,971.40
$36,971.40
$33,406.21
$33,406.21
$34,296.48
$33,406.21
$34,296.48
$33,406.21
$61,158.14
$61,158.14
$61,158.14
$61,158.14
$61,158.14
$57,573.97
$57,573.97
$57,573.97
$57,573.97
$57,573.97
$57,573.97
$56,812.67
$56,812.67
$56,812.67
$56,812.67
$56,812.67
$53,178.57
$53,178.57
$63,103.77
$53,178.57
$63,103.77
$53,178.57
$107,018.79
$107,018.79
$107,018.79
$105,601.62
$105,601.62
$105,601.62
$105,601.62
$126,332.16
$126,332.16
$126,332.16
$125,061.19
$125,061.19
$125,061.19
$125,061.19
$202,000.29
$202,000.29
$202,000.29
$200,310.98
$200,310.98
$200,310.98
$200,310.98
$178,826.93
$178,826.93
$178,826.93
$177,555.96
$177,555.96
$177,555.96
$177,555.96
$327,576.62
$327,576.62
$327,576.62
$326,323.36
$326,323.36
$326,323.36
$326,323.36
$272,270.21
$278,728.58
$278,728.58
$272,270.21
$276,322.75
$276,322.75
$276,322.75
Total cost with

deck
$77,808.54
$79,116.54
$75,516.54
$71,434.14
$67,834.14
$74,265.38
$79,871.67
$72,173.38
$72,189.27
$64,490.98
$78,875.40
$80,183.40
$76,783.40
$72,501.00
$69,101.00
$75,310.21
$77,508.48
$73,218.21
$69,826.08
$65,535.81
$113,538.14
$115,173.14
$110,873.14
$105,570.14
$101,270.14
$109,953.97
$111,588.97
$107,288.97
$101,985.97
$97,685.97
$109,192.67
$110,827.67
$106,527.67
$101,224.67
$96,924.67
$105,558.57
$117,118.77
$102,893.57
$107,515.77
$93,290.57
$180,350.79
$182,639.79
$169,195.59
$178,933.62
$181,222.62
$167,778.42
$199,664.16
$201,953.16
$188,508.96
$198,393.19
$200,682.19
$187,237.99
$296,284.29
$299,227.29
$281,941.89
$294,594.98
$297,537.98
$280,252.58
$273,110.93
$276,053.93
$258,768.53
$271,839.96
$274,782.96
$257,497.56
$432,336.62
$435,606.62
$416,400.62
$431,083.36
$434,353.36
$415,147.36
$377,030.21
$386,758.58
$367,552.58
$381,082.75
$384,352.75
$365,146.75
Cheapest option Deck % of total

with deck
cost
53.9%
54.6%
52.5%
49.7%
47.1%
$64,490.98
56.4%
54.1%
55.2%
49.2%
49.8%
$65,535.81
$97,685.97
$93,290.57
$167,778.42
$187,237.99
$280,252.58
$257,497.56
$415,147.36
$365,146.75
53.1%
53.9%
51.8%
49.0%
46.5%
55.6%
55.8%
54.4%
50.9%
49.0%
46.1%
46.9%
44.8%
42.1%
39.6%
47.6%
48.4%
46.3%
43.5%
41.1%
48.0%
48.7%
46.7%
43.9%
41.4%
49.6%
46.1%
48.3%
41.3%
43.0%
40.7%
41.4%
36.7%
41.0%
41.7%
37.1%
36.7%
37.4%
33.0%
37.0%
37.7%
33.2%
31.8%
32.5%
28.4%
32.0%
32.7%
28.5%
34.5%
35.2%
30.9%
34.7%
35.4%
31.0%
24.2%
24.8%
21.3%
24.3%
24.9%
21.4%
27.8%
27.9%
24.2%
27.5%
28.1%
24.3%
Appendix B: Available plate dimensions

B1
New Zealand-made plates

AS/NZS 1594 Coil plate
Grade
AS/NZS 3678 Plate

Thickness
16
18
20
22

LengthxWidth (mm)
NZS Non Standard Thickness
5
6
8
10
12
14
25
28
32
40
6000x1230

6000x1530

9000x1230
G300Mod*

9000x1530

12000x1230
NA
NA
NA
NA
12000x1530

NA
NA
NA
NA
Mill Lead Time 6 weeks, all thicknesses

6000x1230

6000x1530

9000x1230
G350

9000x1530
12000x1230

NA
NA
NA

12000x1530
NA
NA
NA

6000x1230

6000x1530
G300Mod L0

9000x1230
and L15
9000x1530

12000x1230
NA
NA
NA
NA

12000x1530
NA
NA
NA
NA

6000x1230

6000x1530
G350 L0 and
9000x1230

L15

9000x1530

12000x1230
NA
NA
NA
NA

12000x1530
NA
NA
NA
NA
NOTES:
Lengiths in range <10m for coil plate and between 2.4m and 13m for AS/NZS 3678 plate, in minimum order quanity.
Mill minimun order item quantity for AS/NZS 1594 Coil Plate is the outturn of a slab (approx 14t, 1230 wide and 18t, 1530 wide)
Mill minimun order item quantity for AS/NZS 3678 Plate is approximately 6 tonne.
10mm L0 and L15 tested on subsize Charpy test piece
*When the new AS/NZS 3678 standard is published the mod requirement due to vanadium content will not be required
B2
Grade
45
50
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Commonly imported plates

LengthxWidth (mm)
2400x1220
2700x2400
3000x1520
3100x2400
3400x2400
3600x1520
3600x1800
G250/ G300
4000x2400
5200x2400
6000x1520
6000x1800
6000x2400
7600x2400
9000x2400
Contact the Steel Merchant for expected Lead Time
6000x2000
G350
6000x2400
Contact the Steel Merchant for expected Lead Time
10
12
16
20
25
Thickness
32
40
50

60
65
80
100
120
130
150
200
154
Appendix B: Available plate dimensions
Note that other plate thicknesses (up to 400 mm), dimensions (width in 100 mm increments up to 5m
wide and length up to 25m) and types (weathering steel, L0, L5 and L15, etc) are available. Contact the
steel merchant and/or fabricator for expected lead times. Minimum tonnages may apply.
155
Appendix C: Initial sizing of main girders

The following very basic guidance is offered to give a rough first estimate of girder size for a multi-girder
bridge using plate girders. It is intended only for use where the basic configuration has already been
selected and does not give any indication of the bracing needed to stabilise the girders during
construction or in the final condition. It allows the process of analysis, verification and refinement to
commence; it is not intended for determining quantities for cost estimation.
Reference to loads should be taken to be the design values, ie after the appropriate load factors on actions
have been applied.
At an intermediate support, the maximum shear will occur when a normal load (HN) axle load is
positioned in the lane directly over a girder, with the other axle load located on the longer span side. The
adjacent lane should have the overload (HO) axles in a similar position. The shear can then be estimated
as the sum of components from the HN and HO loads and accompanying UDL load over the span (on the
longer side). For the first two axle loads, use the following factors representing the proportion of load
carried at the end support of a simply supported span, both longitudinally and transversely (eg for a HN
axle load 60% for its position along the span and 90% for a lane that is almost directly above the girder,
giving a shear of 54% of the weight of the HN axle load being carried by the girder). For the UDL loads, use
70% of the load in the span times the proportion for lane position.
The maximum moment at an intermediate support occurs with a different distribution of load but the
worst values can be assumed to coexist. Position the HN axle loads straddling the support and determine
the moment at the end of the longer span (with half the HN axle load on it) assuming it to be a fixed
ended span; multiply by a lane position factor, as above. For the HO axle load, position it at way into the
longer span and determine the fixed end moment; multiply by the lane factor. For the accompanying UDL,
determine the fixed end moment in the longer span and multiply by the lane position factor.
Size the web adjacent to the support so that it can carry 150% of the governing shear (the reserve is
valuable in carrying dead load and contributing to bending resistance). If the bottom flange is inclined, it
will carry some of the shear and the web can be reduced in thickness accordingly.
Determine a force in the bottom flange adjacent to the support by dividing the total moment by the
distance between the flange and the slab. Size the bottom flange so that the stress is 80% of the yield
strength (the 20% allows for dead load and a small reduction for slenderness). Choose a top flange that
has an area of 60% of the bottom flange.
In mid-span, determine moments due to the same three components, with the HN and HO axle loads at
mid-span; consider the span as simply supported. Consider that 40% of the total moment is carried by a
single girder (this makes some allowance for continuity and assumes transverse sharing between girders).
In mid-span, provide a web that has an area of 60% of that at supports and size the bottom flange to carry
a force of the total moment divided by the depth between the flange and the middle of the slab, at a stress
level of 95% of the yield strength. Choose a top flange area that is 80% of that of the bottom flange (this
will be needed for stability during construction).

Compression flanges should always be proportioned so that they are at least non-compact (for grade 350
steel, limit the outstand/thickness ratio to 11.2). Tension flanges should be limited to an outstand/
thickness ratio of 20, for robustness (but if they could go into compression during construction, comply
with the non-compact limit).
156
Appendix D: Background to the fatigue design criteria for bridges
Appendix D: Background to the fatigue design

criteria for bridges
Taken from Recommended draft fatigue design criteria for bridges (Clifton 2007a).
The fatigue loading in AS 5100.2 (SA 2004) has been developed to be consistent with the M1600 traffic
design live loading. It is based on the same family of future vehicles determined for Australia. In applying
it to New Zealand, expert assessment by Dr Clifton has been made to amend the provisions suitable for
New Zealand conditions.
Fatigue is the process of cumulative local structural deterioration of a component or connection when
subjected to repeated, fluctuating stresses. In bridges, it is normally caused by the repeated loading of a
structure by moving vehicles and occurs over an extended period of time. While fatigue damage in main
structural members is of principal concern, distortion or distortion restraint induced fatigue damage has also
been reported in a number of instances in secondary bracing members or their connections (Fisher 1977).
Fatigue damage is commonly exhibited as cracks or fracture of the component material and usually occurs
at locations of stress concentration. The fatigue criteria presented below and discussed in AS 5100.5 and
AS 5100.6 (SA 2004), have been developed with the intention of limiting the development and propagation
of cracks under repeated loading to prevent fracture during the life of the structure. The fatigue behaviour
of steel structures, including composite steel and concrete structures, is relatively well understood and
direct reference may be made to current international best practice and standards. The fatigue behaviour
of concrete is not as well defined. Reference has been made to Eurocode and recent Swiss research.
Reference should be made to the AS 5100.5 and AS 5100.6 commentaries for further details of materials
aspects of the fatigue criteria.
In developing the fatigue design criteria, a fatigue life of 75 years has been assumed. This approach is
compatible with a 100-year design life on the basis that:
Bridges will be inspected regularly and that intervention will occur when fatigue damage is detected.
This process should ensure that fatigue damage is controlled and this has been shown to be the case
in practice.
The uncertainty associated with the prediction of the fatigue life is such that only a relatively low
percentage of bridges for which the theoretical design fatigue life has been reached actually exhibit
fatigue damage.
Initial work on developing fatigue loading criteria concentrated on the use of specific vehicles to represent
the range of typical future freight vehicles. This approach was replaced in AS 5100.2 by the use of the
A160 and M1600 truck load for Australia and New Zealand to remove the requirement to introduce further
design vehicle loads. In the New Zealand application this involves having to use an axle load case specific
for fatigue assessment. This is unavoidable and is consistent with future intentions to align the
New Zealand and Australian bridge loadings and design standards more closely. In both instances,
assumptions have been made about the rate of increase in freight task and corresponding increase in
heavy vehicle volumes and mass together with assumptions about commercial vehicle configurations on
four principal route types.
When determining the basic fatigue design loads for New Zealand, the recommendations from Beamish et
al (2006) and subsequent discussions with the authors is that the Australian fatigue design provisions are
appropriate for New Zealand use (based on a limited number of weigh stations) but with the route factors
157
given for the closest equivalent Australian class of road used and multiplied by 0.9. This has been done
below.
This approach has removed the need for the designer to determine the number and amplitude of the
stress cycles due to the passage of actual trucks. In reality the passage of a single vehicle over a bridge
produces a number of cycles of stress of varying amplitude. The cumulative fatigue damage of this stress
history is converted to the equivalent damage of one cycle of amplitude (f*equiv)m, where:
n f *
i
f
*
equiv
(Equation D.1)
equiv
Where:
nequiv
= 1.0.
The design cumulative damage caused by the passage of the assumed mix, mass and number of vehicles
is expressed as an equivalent number of cycles of 0.70(A160 axle load) (1 + ), in the case of very
short spans or 0.95(HN moving traffic load without UDL) (1 + ).
The corresponding damage caused by a single passage of the HN truck without UDL or the A160 axle load
is (M1600)m or (A160)m.
Thus the design fatigue damage can be expressed as
nA160(f*
)m or
nM1600(f*
0.7A160
)m, where:
0.7M1600
nA160
is the number of fatigue stress cycles for the A160 axle load
nM1600
is the number of fatigue stress cycles for the M1600 lane load
(f*
= equivalent cumulative fatigue function for 0.70(A160 axle load) (1 + )
0.7A160
(f*
)m
0.7M1600
)m
= equivalent cumulative fatigue function for 0.70(M1600 moving traffic load without UDL)
(1 + ).
The design number of fatigue cycles is applicable to both simply supported, continuous and cantilever
spans.
In determining the maximum stress range in the component under consideration, only 70% of the effect of
the basic load is used. This reduction has been introduced to take the following into consideration:
The actual stresses in a component are generally less than the theoretically calculated values because
of alternative load paths, (such as bridge barriers) and the magnitude of actual components in
comparison with line elements used to represent them in analysis;
The actual lateral position of heavy vehicles varies and does not generally coincide with the critical
lateral position.
In general terms for Australia, the mass of heavy vehicles is assumed to increase from current values to
that equivalent to SM1600 vehicles over a period of about 50 years. Emphasis is placed on the average
mass of vehicles rather than maximum allowable mass as the percentage of vehicles loaded to maximum
allowable mass may be expected to decrease as the allowable mass is increased over time and volume
tends to control. The compound rate of increase of freight is assumed to be about 5% on major interstate
routes down to only about 1% on lesser routes. A saturation volume of between 4000 and 4500 heavy
vehicles per lane per day is also assumed. These assumptions are made for New Zealand.
For the determination of the number of fatigue stress cycles, a route factor has been introduced to take
into consideration the average mass of heavy freight vehicles on different types of routes. In adapting this
for New Zealand, the classes of roads used here are specified from the Bridge manual and the route
factors used in AS 5100 (SA 2004) are reduced by a factor of 0.9. Vehicles travelling on national highways
158
Appendix D: Background to the fatigue design criteria for bridges
and motorways are generally maximum sized legal vehicles, fully laden and include a high proportion of
medium vehicles. Vehicles travelling on major urban roads are more likely not to be fully loaded, with
many empty. Vehicles on other regional roads are less likely to be maximum sized than those on the
national road network. Vehicles on urban roads other than motorways are likely to be shorter and lighter
than heavy vehicles on other types of roads. Considerable use has been made of weigh-in-motion data
from Australia in determining the different route factors and limited calibration made with weigh station
data from New Zealand.
General traffic data has been used for Australia to determine the presence of heavy vehicles in different
lanes of multi-lane roads. This data indicates that for rural roads more than 95% of all heavy vehicles
travelled in the slow lane. For urban roads, particularly for freeways with frequent on and off ramps, a
maximum of about 65% of heavy vehicles travelled in a single lane. This data has been used in specifying
the number of heavy vehicles per day per lane to be used based on the total number of heavy vehicles
travelling in that direction. The applicability of this for New Zealand needs to be confirmed or
recommended modifications made.
Guidance on the number of trucks per lane for the different classes of road are available from the NZTA in
New Zealand for the national roads.
The final aspect is the source of the dynamic load allowance. This is specified in table 6.7.2 of AS 5100.2
for the Australian loads.
159
Appendix E: Single span ladder deck bridge

E1
Design statement
The bridge carries a two-lane (each 3.5m wide) single carriageway road over a flood plain.
The carriageway has two 1m wide shoulders and has 2m wide footways on either side.
Speed environment is signed as a maximum of 50km/h but a design threshold of 60km/h is used.
For barrier design the average annual daily traffic (AADT) is 15,000 vehicles per day with 20%
commercial vehicles. An offset of 1.2m is assumed. Thus, a G9 modified TL4 Thrie beam is
considered.
The bridge clearance is minimum 1.5m from the soffit to design flood level, therefore flood load does
not need to be considered.
The bridge is to be designed in accordance with following standards:
Bridge manual (NZTA 2012)
AS 5100.6 (SA 2004)
NZS 3101 (NZS 2006).
Structural material properties:
structural steel: grade 300
concrete: C40
steel reinforcement: grade 500.
Environmental exposure classification for superstructure = A2 inland, exterior.
Density of steel = 78.5kN/m3.
Density of reinforced concrete = 25kN/m3.
Nominal thickness of surfacing = 130mm.
Density of surfacing = 23kN/m3.
Steel parapet selfweight = 2kN/m for both TL4 and pedestrian barriers.
The construction load is 0.75kN/m2 and the weight of the temporary formwork is assumed to be
0.5kN/m2. Additionally, wet concrete is assumed to have a density of 1kN/m3 greater than that of
hardened concrete.
Pedestrian live load = 5kPa. This value can be reduced when considered in conjunction with road
traffic loading.
Draft fatigue design criterion for bridges (Clifton 2007a) is recommended for fatigue assessment.
Coefficient of linear thermal expansion = 11.7x10-6 per C.
Characteristic value of shrinkage is sh 0.0002
160
E2
Structural arrangement
The bridge carries a two-lane single carriageway road over a flood plain. The carriageway has 1m wide
shoulders and has a 2m wide footway on either side. A ladder deck girder arrangement has been chosen
and a deck slab of 250mm has been assumed. The deck cantilevers 1.3m outside the centrelines of the
outer girders. The bridge is assumed to have 24.5m span with pinned end supports.
Figure E.1
E3
Ladder deck bridge structural arrangement
Design basis
The bridge is to be designed in accordance with the Bridge manual, AS 5100.6 and NZS 3101.
The fatigue limit state is verified for the reference stress range due to the load application based on
Clifton (2007a).
Crack widths in the deck slab are verified at the serviceability limit state (SLS) based on NZS 3101.
E3.1
Load combinations
Factors and load combinations of actions are given in the Bridge manual, tables 3.1 and 3.2 for the SLS
and ultimate limit state (ULS) respectively.
E3.2
Factors on strength
The values of various capacity reduction factors ( ) for strength limit states are given by AS 5100.6, table
3.2 for steel, and NZS 3101, clause 2.3.2.2 for concrete.
161
E3.3
Structural material properties
It is assumed that the following structural material grades will be used:
concrete: C40
For structural steel, the value of
fy depends on the product material

ASI design
For rolled sections use:
capacity
320MPa for t 11mm; 300MPa for 11mm <t 17mm; and 280MPa for t
tables
>17mm. For plates use:

320MPa for t 8mm; 310MPa for 8mm <t 12mm; 300MPa for 12mm <t 20mm; and 280MPa for t
>20mm.
Note that designers may wish to use grade 350 steel for better economy.
'c
For concrete, f = 40MPa

For steel reinforcement, fy = 500MPa
The modulus of elasticity for both structural steel and steel reinforcement is taken as Es 200GPa
The modulus of elasticity of the concrete is given by NZS 3101 as:
E =
6900
c
1.5
'
3320 f +
2300
NZS 3101,
clause 5.2.3
Ec = 32GPa
This 28-day value will be used for determination of all short-term effects and resistance and the modular
ratio is thus:
nS = 200 / 32 = 6.25
For long-term effects, the modular ratio is:

E
Ec
=
c,L
32
= 10.50
1+ 2
1+
cc
nL = 200 / 10.50 = 19
The conservative value of
cc is given in National Roads Board (1984) Road Research Unit bulletin 70 (RRU
70) as 2.00.
E3.4
Durability requirements
E3.4.1
Concrete
Environmental exposure classification

Superstructure: A2 inland, exterior
Minimum cover to reinforcement

Surfaces in contact with the ground = 75mm
Surfaces with a damp-proof membrane between the ground and the concrete = 50mm
162
Elsewhere = 40mm.
E3.4.2
Steel
For an appropriate and cost-effective coating system for structural steelwork, the New Zealand steelwork
corrosion and coatings guide (El Sarraf and Clifton 2011), is used in conjunction with AS/NZS 2312 (SNZ
2002) Guide to the protection of structural steel against atmospheric corrosion by the use of protective
coatings and NZS 3404.1 (SNZ 2009).
E4
Action on the bridge
E4.1
Permanent actions
E4.1.1
Self-weight of structural elements
The density of steel is taken as 78.5kN/m3 and the density of reinforced concrete as 25kN/m3. Selfweights are based on nominal dimensions.
E4.1.2
Self-weight of surfacing
The nominal thickness of the surfacing is 130mm. Assume that the density is 23kN/m3. The self-weight
generally produces adverse effects and as a result is based on nominal thickness + 55%. This follows
international practice; however, New Zealand practice differs and reference should be made to the Bridge
manual, section 3.4.2.

DL 1.550.1323 4.63kPa
E4.1.3
Self-weight of footway construction
The nominal thickness of the footway comprising a concrete slab is 250mm and a uniform density of
25kN/m3 is assumed. The self-weight is based on the nominal dimensions.
E4.1.4
Self-weight of parapets
A nominal value of 2kN/m is assumed for each parapet.

Note that this value will be increased if a solid concrete crash barrier is used.
E4.2
Construction loads
For global analysis, a uniform construction load of 0.75kN/m2 is assumed during casting. The use of
permanent precast planks is assumed and thus there is no extra load for formwork. Additionally, wet
concrete is assumed to have a density of 1kN/m3 greater than that of hardened concrete; for a slab
thickness of 250mm this adds 0.25kN/m2.
The total construction load is thus: CN 0.75 0.25 1.0kPa .
E4.3
Traffic loads
E4.3.1
Road traffic
For the road carried by this bridge, the Bridge manual specifies the HN-HO-72 traffic loading but only
HN72 has been considered here for simplicity.

E4.3.2
Pedestrian traffic
Pedestrian traffic is represented by the value in the Bridge manual, section 3.4.14(b).
FP 5kPa
163
This value can be reduced when considered in conjunction with road traffic loading.
E4.3.3
Fatigue load
Clifton (2007a) is recommended for fatigue assessment.
E4.4
Thermal actions
E4.4.1
Overall temperature change
Refer to section
6.4.2 in the main
part of the guide
For a change of length in composite sections, the coefficient of linear thermal expansion is 11.7x10-6 per
C. This is then used for determining soil pressure on integral bridges or determining the expansion
length in the case of simply supported abutments.

E4.4.2
Differential temperature change

Refer to Bridge
The vertical temperature difference given in the Bridge manual, section 3 is used and
manual, section
the temperature difference will be considered to act simultaneously with the overall
3.4.6
temperature change.
The effects of vertical temperature gradients are derived for the positive differential temperature
conditions (where the top surface is hotter than the average temperature of the superstructure) and for he
negative temperature differential conditions (where the top surface is colder than the average temperature
of the superstructure).
Note: The negative temperature variation to be considered is the same as that for bridge type 1 from
figure 17.3 of AS 5100.2.
Figure E.2
E4.5
Temperature variation with depth
Seismic actions
Section 5 of the Bridge manual is used to determine the horizontal force generated
by the seismic action on the bridge superstructure. These forces are transferred into
the supports (piers and abutments), which are designed to resist this horizontal
Refer to section
6.5 in the main
part of the guide
force (as shown in figure 5.4 of the Bridge manual). Therefore, seismic actions acting on the
superstructure are ignored as they are minimal as long as the load path is available to transfer these
actions into the substructure.
164
E5
Girder makeup and slab reinforcement
E5.1
Main girder
Figure E.3
Main girder arrangement
24.50m span girder

Top flange
600 x 25
Web
20
Bottom flange
800x 50
Top rebar
HD20-150
Bottom rebar
HD20-150
The overall girder depth is 1350mm. Cover to the centroid of the top and bottom reinforcement is 55mm.
This complies with NZS 3101.
Cross girders are positioned at 3500mm centres and are connected to the main girders by bolting to flat
transverse web stiffeners.
One alternative configuration to an in-situ concrete slab is to include precast planks, which typically have
prestressing in the bottom, and an in-situ slab with lighter reinforcement.
E5.2
Figure E.4
Cross girders
Cross girder arrangement
Overall depth 750mm at the ends, 896mm at the centre

Flanges: 300 x 25
Web:
16mm
The web is unstiffened, except possibly at the cross girder mid-span if the cross girders need to be
braced.
165
E6
Beam cross section
E6.1
Section properties main girders
Figure E.5
Refer to section
7.1.1 in the main
part of the guide
Slab effective width
For determination of stresses in the cross section and resistances of the cross section, the effective width
of the slab, allowing for shear lag is needed. The following calculations summarise the effective section
properties for the section considered.
The equivalent span for effective width is:
Refer to
Lc 24500mm
AS 5100.6,
beff 1300 min
L c S
, , 6t
10 2
24500 11700
1300 min
10
clause 6.1.7
2800mm
, 6 250

Refer to
AS 5100.6,
Properties for the gross section in the mid-span are tabulated below.
clause 4.4.1
Table E.1
Bare steel cross sections

Span girder
Unit
80,500
mm2
ENA
479
mm
2.4E+10
mm4
Section modulus, top flange
Zx, tf
27E+06
mm3
Section modulus, bottom flange
Zx, bf
50E+06
mm3
Area
ENA height
Second moment of area
Section class
Not compact
M
Bending moment capacity
sx
7560
kN.m
Slenderness check:
e,t_ flange
e,t_ flange
bf tw
2t f
fy,flange
250
600 20 280
2 25 250
12.28
Refer to
AS 5100.6,
clause 5.1.2
166
e,web
e,web
fy,web
d1
tw
250
1275 300
20
69.8
250
e
14
y,flange
(1350 479 25)
re
0.66
1275
60
60
e
91
y,web
re 0.66
e,t _ flange
e
table 5.1
0.88
14
69.8
0.77
91
y,web
e,flange
e
12.28
y,flange
e,web
e
Refer to AS 5100.6,
y,flange
e,web
e
flange governs
y,web
e e
12.28
S
flange
e 8
p
e e
p
S
Therefore, the bare steel section is not compact.

Table E.2
Composite cross section (short term) sagging (nS=6.25)

Span girder
Unit
192,500
mm2
ENA
1059
mm
Ix
7.1E+10
mm4
Section modulus, top of slab
Zx, c
131E+06
mm3
Zx, tf
243E+06
mm3
Zx, bf
67E+06
mm3
Compact
22,596
kN.m
Area
ENA height
Section class
Mp
The cross section of the span girder is compact, provided the top flange is restrained by shear connectors
within the spacing limits given in AS 5100.6, clause 6.1.8. (the spacing of the shear stud is more likely to
be 150mm).
Plastic moment capacity for a compact composite section can be calculated using the formula given in
AS 5100.6, appendix E.
fy A
dh
'
0.85fcb
167
d
h
28080500
0.85402800
237mm
dh ds
Mp Afy dg ds dh
dg 1350
2
250
2
479 996mm
250 237 6
10 22, 596kN.m
Mp 28080500 996
Figure E.6
Refer to AS 5100.6,
appendix E
Main girder and slab stress blocks
Note: During the construction of a composite bridge, it is quite likely a beam will change its section class,
because the addition of the deck slab both prevents local buckling of the top flange and significantly
shifts the neutral axis of the section. Typically, a mid-span section could be compact after casting the slab
but not prior to this. As a consequence, checks at intermediate stages of construction should be based on
the relevant classification at the stage being checked.
Table E.3
Composite cross sections (long term) sagging (nL=19)
Area
ENA height
Span girder
Unit
117,250
mm2
ENA
791
mm
4.9E+10
mm4
x, c
61E+06
mm3
x, tf
88E+06
mm3
x, bf
62E+06
mm3
I
Z
Z
Z
168
E6.2
Effects of temperature difference and shrinkage
The primary effects of differential temperature through the depth of the cross section of
Refer to
a member are considered in the design. In addition, the secondary effects in continuous
AS 5100.6,
members, due to redistribution of the moments and support reactions caused by the
clause 4.9.1.1
primary effects are also considered.

Longitudinal stresses and shear forces due to differential temperature effects are calculated by elastic
theory assuming full interaction between the concrete slab and the steel beam. The stiffness is based on
the transformed composite cross section using a modular ratio appropriate to short-term loading and
assuming the concrete slab has an effective width calculated in accordance with AS 5100.6, clause 4.4.1.
When the effects of shrinkage modified by creep adversely affect the structure, they are
calculated in the manner described for differential temperature effects, but using a
modular ratio appropriate to long-term loading. The beneficial effects of the creep of
Refer to
AS 5100.6,
clause 4.9.1.2
concrete are taken into account.

The effects of the temperature difference and shrinkage on design of the composite section are neglected
as they are not adverse in the case of a simply supported bridge. However, designers may wish to check
this. The longitudinal forces due to the primary effects of shrinkage and differential temperature are both
considered in the design of the stud shear connector and transverse reinforcement.
E6.2.1
Temperature difference
For the calculation of primary effects, the short-term modulus of concrete is used.
Ec 32GPa
Es 200GPa
For each element of the section, calculate stress as the strain times modulus of elasticity, and then
determine force and centre of force for that area. The restraint moment in the inner beam, due to the
characteristic values of temperature difference is noted in the Bridge manual as:
o
T (32 0.2d) C
d 50mm
o
T 22 C
The moment release stress is shown diagrammatically in figure E.7.

Figure E.7
Moment release stress over the depth of the span-girder section
169
E6.2.2
Shrinkage
The effects of shrinkage are calculated for the long-term situation.

The characteristic value of shrinkage is sh = 0.0002 and the long-term modular ratio is used.
Table E.4
Temperature and shrinkage axial force

Force
Differential temperature axial fixity force
4053
kN
Shrinkage axial fixity force
3754
kN
E6.3
Section properties cross girders
The gross composite section of the cross girder includes half of the width of the slab to the adjacent cross
girder.
For the effective composite section, taking account of shear lag, the cross girder is effectively simply
supported and thus Lc 11700mm and the effective width of slab is:
Refer to AS 5100.6,
11, 700
b
min c ; S;12t min
; 3500;12 250
s
eff
2340mm
5
clause 6.1.7
Refer to AS 5100.6,
Assume there is only a single row of connectors on the beam centreline.
clause 4.4.1
The effective section properties of the cross girder at mid-span are tabulated below.
Figure E.8
Cross girder and slab effective width
Table E.5
Cross girder section properties

Bare steel
Short-term
Long-term
Unit
28,536
122,136
59,249
mm2
ENA
448
887
745
mm
3.6E+09
1.1E+10
8.7E+09
mm4
Zx, c
44E+06
22E+06
mm3
Zx, tf
8E+06
1256E+06
57E+06
mm3
Zx, bf
8E+06
13E+06
12E+06
mm3
Compact
Compact
Compact
Msx
2688
5176
kN.m
Area
ENA height
Section class
The cross section of the span girder is compact, provided that the top flange is restrained by shear
connectors within the spacing limits given in AS 5100.6, clause 6.1.8.
170
E7
Global analysis
E7.1
Model
Since the bridge is simply supported at its ends, a hand calculation effectively provides sufficient design
actions at the main and cross girders.
Note, if a sophisticated computer model is used, torsional properties need to be considered.
E7.2
Construction stages
The whole deck span will be concreted in one stage. The edge beams will be concreted afterward.
Separate analytical models are therefore provided for:
Stage 1: All steelwork, wet concrete
Stage 2: Composite structure (long-term properties), the weight of the edge beams is applied
Stage 3: Composite structure (short term properties).
E7.3
Analysis results
The following results are for design values of actions, ie after application of appropriate factors on the
characteristic values of actions.
For construction loading, results are given for the total effects at each construction stage. For traffic
loading the results are given for the combination of traffic and pedestrian loading for worst bending
effects at the central span location.
E7.3.1
Stage 1
Selfweight of steelwork
Selfweight of concrete
Construction load
Table E.6
Stage 1 design actions

ULS
Distance from support (m)
F z (kN)
M y (kN.m)
F z (kN)
1046
766
12.25
6410
4690
Load combination 5B is assumed to be critical
E7.3.2
SLS
My (kN.m)
Stage 2
Selfweight of parapets
Selfweight of carriageway surfacing
Selfweight of footpath construction
Removal of construction loading
171
Table E.7
ULS
SLS
My (kN.m)
F z (kN)
M y (kN.m)
F z (kN)
521
386
3190
2362
0
12.25
Note: Load combination 5B is assumed to be critical
E7.3.3
Table E.8
Stage 3
HN72 traffic loads for worst sagging at mid-span and worst shear at supports
ULS
SLS
My (kN.m)
F z (kN)
M y (kN.m)
F z (kN)
1182
703
12.25
6836
4063
Note: Load combination 1A is assumed to be critical
Please note that in this case the maximum sagging moment at mid-span occurs when the HN72 axle loads
are positioned at 8.5m and 13.5m respectively from the supports.
E8
Design values of the effects of combined actions
Design values are given in section E8 for certain situations in the design of the main girder and cross
girder beams. In practice, the design of other parts of the structure would also need to be considered.
E8.1
Effects of construction loads (ULS)
Generally, the effects of construction loads apply to different cross-section properties. The cross sections
for the main girder beams are different at stages 1 and 2. The following tabulations summarise the forces
and moments at each stage and the stresses due to those effects, for selected cross sections. According to
the Bridge manual the load combination 5B is the most critical during the stages of construction.
Table E.9
Stress at mid-span due to construction loads

Bottom flange
M
x, bf
Top flange
x, tf
Top of slab
kN.m
kN
106mm3
MPa
106mm3
MPa
Stage 1
6410
1046
50
128
27
-237
Stage 2
3190
521
62
51
88
-36
9600
1567
179
-273
x, c
106mm3
MPa
61
-52
-52
E8.2
Effects of traffic load plus construction loads (ULS)
Effects due to traffic actions are determined from the short-term composite section in mid-span.
172
Table E.10
Stress at mid-span due to traffic load plus construction loads

Bottom flange
Top flange
My
Fz
Zx, bf
kN.m
106mm3
MPa
Construction
9600
1567
Traffic HN72
6836
1182
16,436
2749
Zx, tf
Zx, c
106mm3
MPa
106mm3
MPa
179
67
Top of slab
-273
102
243
281
-28
-52
133
-301
-51
-103
The resulting stresses in the tables above indicate that the steel section remains elastic under construction
loading and is plastic under a full service load. These results indicate an economical section size.
E8.3
Effects in intermediate cross girders
With the traffic load positioned for the worst effects on the cross girders, the worst sagging occurs in the
middle cross girder.
E8.3.1
Worst sagging on cross girder (ULS)
The following stresses are elastic stresses on the effective cross section allowing for shear lag.
Table E.11
Cross girder stress at mid-span

Bottom flange
Top flange
Top slab
My
Fz
Zx, bf
kN.m
kN
106mm3
MPa
106mm3
MPa
Stage 1
650
222
79
-79
Stage 2
287
98
12
24
57
Construction
937
320
Traffic HN72
1998
460
2935
780
Zx, tf
Zx, c
106mm3
MPa
22
-13
103
13
154
5
-84
1256
257
-13
44
2
-86
-46
-59
Values for SLS would be needed if the total stresses exceeded the design elastic values but, by inspection,
they are not exceeded.
E9
Verification of bare steel girder during construction
The two main girders are susceptible to lateral torsional buckling under the weight of
the wet concrete (ie before it hardens and provides restraint to the top flanges). The
beams are partially restrained against buckling by the presence of the cross girders.
The cross girders provide flexible torsional restraint to the beams.
Refer to figure 7.4

in the main part of
the guide
173
Figure E.9
E9.1
Lateral torsional buckling of the main girders
Restraint to main girders
To provide restraint to the main girders, the cross girders should be designed for a situation where the
two critical flanges try to move towards each other and put the cross girder into compression. Therefore,
AS 5100.6, clause 8.4.2 states that the transverse force acting on the lateral restraint is taken as 2.5% of
the most onerous ULS force in the critical flange in the main girder. However, for ladder deck bridges the
transverse member loads may be more significant. Hence the following check may be undertaken
Stress at top flange under construction load = 237MPa
Maximum flange force (under construction loading) = 2376002510
3555kN
2.5% flange load = 0.0253555 89kN

This load is resisted by the cross girder axially. The following calculation is conservatively based on the
minimum height of the cross girder section.
N* Nus
N* Nuc
Refer to AS 5100.6,
Nus = Ns
Ns k f An fy
clause 10.1
k f Ae Ag
clause 10.2.1
e,flange
e,flange
Refer to AS 5100.6,
bf tw
fy,flange
2t f
250
300 16 280
2 25 250
6.0
174
e,web
fy,web
d1
tw
250
700 300
48.0
e,web
16 250
From AS 5100.6, table 10.2.4
ey,flange 14
ey,web 115
Therefore,
b f t w ey,flange b t
f w

be,flange
2 e,flange 2
300 16 14 300 16
6
2
2
300 16
be,flange
142mm
2
ey,web
be,flange
d1
e,web
115

be,web 700
700
48
be,web
d1
be,web 700mm
2
A 2300257001626,200m
m
e
2
A 2300257001626,200m
m
g
k f A e Ag 26,200 26,200 1.0
3
7336kN
Ns k f Anfy 26,20028010
Nus = Ns = 0.9 7336 = 6600kN
Also,
Nuc = Nc
Nc = cNs Ns
6
4
Iy 11310 mm
ry
Iy
A
Le

r
y
11310
26200
6
66mm
fy 11700
300
kf
1.0 194
250
250 66
2100
13.5
a =
Refer to AS 5100.6,
clause 10.3.3
2100(194 13.5)
=
= 10.32
15.3 + 2050
1942 15.3 194 + 2050

175
b 0.5
ab 194 10.32 0.5 199
0.00326 199 13.50
0.6

2+1+
= 90
2 2
=
18
+ 1 + 0.6
90
189
2 2
90
90
= 0.68
90
90
2
2
c 1 1 0.68 1 1
0.195
0.68 189
c can also be directly read off table 10.3.3 in AS 5100.6
3
Ns k f Anfy 1.026,20028010 7336kN
N
=
N
=
0.195
7336
=
1431kN
c
c
s
Nuc = Nc = 0.9 1431 = 1288kN
N* Nuc OK
N* Nus OK
Also,
*
Mx
*
6410kN.m
Refer to AS 5100.6,
Mx Mrx
N*
Mrx = Msx 1
Ns
M = 0.9 7560
1-
clause 11.3.2
89
= 6712kN.m
rx
0.9 7336
*
Mx = 6410kN.m Mrx = 6712kN.m OK
E9.2
Refer to AS 5100.6,
clause 5.6.1
Verification
Maximum sagging bending moment at the mid-span under construction load is 6410kN.m.
The flexural capacity of the main girder is given in AS 5100.6, clause 5.6 as follows:
Mb = msMs Ms
At the mid-span the bending moment assumed to be constant, therefore, m 1.13

Le = ktkLkrL
kt = 1.0
kL =
1.4
kr = 1.0
(LL)
Le = 3.5 1.4 = 4.9m

E = 200,000MPa
176
G = 80,000MPa
9
4
m
10
m
I = 2.58
y
Iw = 0
J = 3.92
7
10
4
m
Ms = 7560kN.m
2
2
EIy EIw
2 GJ
2
Le
Le
Mo
=
2
2
200,000 2.58
M
=o
7
9
10
4900
80,000 3.92 10
-6
10
Mo = 25,791kN.m
Ms Ms
= 0.6
+ 3 -
Mo Mo

7560 2 7560
= 0.6
+ 3 = 0.88
25,791
25,791
M = 1.13 0.88 7560 = 7518kN.m M =

s
7560kN.m
b
Mb = 0.9 7518 = 6766kN.m

M* = 6410kN.m < M = 6766kN.m - OK
b
Refer to AS 5100.6,
Also,
clause 5.10.2
V* Vv
dp
tw
1275
20
= 63.75
82
fy
300
250
250
Vv = Vw
82
= 1275 20 = 25,500m
m
w
= 74.8
-3
Vw = 0.6fy Aw = 0.6 300 25,500 10 = 4590kN
Vv = 0.9 4590 = 4131kN

V* = 1046kN Vv = 4131kN - OK
E10 Verification of composite girder
E10.1 Sagging bending in main girder

The elastic design bending resistance for a beam constructed in stages depends on the design effects at
the stages.
177
The bare steel section is not compact in bending and the composite cross section is compact. The effects
(stresses) in the cross section have been calculated on the basis of gross section properties for effects on
the steel beam plus effective section properties on the composite section.
The composite cross section is compact (PNA in the top flange) so the plastic resistance can be utilised.
The plastic bending resistance of the short-term composite section is 0.9 x 22,596=20,336kNm and the
total design value of bending effects is 16,436kNm. The cross section is satisfactory by inspection. It can
also be seen that the stresses calculated elastically, taking account of construction in stages, are
satisfactory, as follows:
Figure E.10
Elastic stresses in the girder
E10.2 Verification of cross girders

As previously noted, the stresses in the cross girders are within elastic limits for the loads considered in
the analysis. However, these transverse girders are required to prevent buckling of the slab where it is in
compression. The cross girders need to be both stiff enough and strong enough to perform this function
in addition to the resistance to the effects already calculated.
Note that the Intermediate cross girders effectively act as simply supported beams in carrying the loading
from the slab
The plastic bending resistance of the short term composite section is 0.9 x 5515=4964kNm and the total
design value of bending effects is 2935kNm. The cross section is satisfactory by inspection. In addition,
the combined axial and bending actions should be checked. This would be derived from a detailed deck
analysis.
Note: This design is very conservative, therefore, designers may wish to refine it for economy.
E11 Longitudinal shear
Refer to section 6.1.7 in the

main part of the guide
The resistance to longitudinal shear is verified for the web/flange weld, the
shear connectors and the transverse reinforcement at the supports and at mid-span. (In practice,
intermediate values would also be verified, to optimise the provision of shear connectors.)
178
E11.1
Effects for maximum shear
Table E.12
ULS values at supports

Supports
Shear on steel section (stage 1)

Shear on long term composite section
Shear on short term composite section (worst effects)
Table E.13
1046
kN
521
kN
1182
kN
SLS values
Supports
766
kN
386
kN
703
kN
E11.2 Section properties

To determine shear flows the parameter At yc It is needed for each section and stage.
Table E.14
Section properties
Web/top fl
Top fl/slab
A t y c It
At yc It
m-1
m-1
Steel section
0.537
Long term section
0.682
0.514
Short term section
0.715
0.656
E11.3 Shear flow at ULS

Force at the web/top flange junction
At abutment
1046 x 0.537 + 521 x 0.682 + 1182 x 0.715 =
1762
kN/m
1043
kN/m
Force at the flange/slab junction

At abutment
521 x 0.514 + 1182 x 0.656 =
E11.4 Shear flow at SLS

Force at flange/slab junction
At abutment
386 x 0.514 + 703 x 0.656 =
660
kN/m
The shear flow at SLS is required for verification of the shear connectors.
E11.5 Web/flange weld

Design weld resistance is given in tables 9.8 and 9.9 of ASI (1999) design capacity tables.
For 6mm throat SP fillet weld tw=8mm.
179
nw
6mm
= 1.11kN / mm
Resistance of two welds = 2220kN/m > 1762kN/m, shear flow at top flange OK
Shear flows at bottom flange are slightly less and are OK by inspection.
Refer to section
E11.6 Shear connectors
7.7.3 in the main
Shear stud connectors 19mm diameter, 150mm long are assumed, with f =410MPa.
part of the guide
The resistance of a single stud is given by AS 5100.6, clause 6.6.4.4 as the lesser of:
2
fuc
f0. =
63d
ks
bs
or
2
'
f =
0.
f E
63d
ks
bs cy c
'
fcy
= 40MPa
d =
16mm bs
fcu =
410MPa
'
Ec = 5050 fcy = 5050 40 = 32,000MPa
Note: For the purpose of evaluating the resistance of headed stud connectors embedded in solid slabs, the
'
modulus of elasticity for concrete may be taken to be Ec = 5050 f cy .

From structural reliability analyses conducted by HERA, it has been shown that the
Refer to AS 5100.6.
use of this value produces predictions that better reflect the resistance of studs in
clause 6.6.3.2
physical tests.
Therefore the design resistance of a single-headed shear connector is:
f = min 0.63
ks
2
410;0.63
19
40 32,000= 93.25kN
19
uL uLs
uLs = 0.55nfk s
= 1.0
3754 5
*
kN / m
uL = 660kN / m +
24.5 2
shrinkage PE = 1043kN / m
If studs are grouped and spaced at 150mm spacing along the beam (to suit transverse reinforcement),
then a row of three studs has a design resistance of:
Refer to section
0.55 93.25 3
uLs =
= 1025.75kN /
m 0.15
7.7.7 in the main

part of the guide
E11.7 Transverse reinforcement

Consider the transverse reinforcement required to transfer the full shear resistance of the studs, ie
1025.75kN/m as well as the maximum ULS shear flow, ie 1046kN/m.
180
Figure E.11
Transverse reinforcement
For a critical shear plane around the studs, shown dotted above, the shear resistance is provided by twice
the area of the bottom bars. The design shear resistance of the transverse reinforcement is given by
AS* 5100.6, clause 6.6.5.2 as follows:
Lp 0.9us 0.7Ats fry

and
'
Lp 0.15ufc
Assumed HD20150mm:
s = 1MPa
u = 2 150 + 600 = 900mm

A = 2 314 1000 150 =
4188m
st
/m
f 450MPa
ry

L min 1.00.9 900 1.0
+
*
uLp
0.7 4188 450

;1.00.15 900 40= 2130kN / m
3
10

1.35 3754 5
= 1563kN / m
= 1046kN / m +
24.5
shrinkage PE
uLp = 1563kN / m uL = 2130kN / m - OK
The transverse bars are adequate.
E12 Cross girder to main girder connection

Consider the cross girders at mid-span and adjacent to the support.
E12.1 Structural arrangement of connection

The connection between main girders and cross girder is designed for the bending moment determined
from 2.5% of the moment-induced compression force in the top (critical) flange of the main girder. This
acts at a lever arm corresponding to the vertical distance between the centroid of the bottom flange (ie the
tension flange) and the centroid of the bottom flange. This moment has to be induced in the connection
design in conjunction with the critical shear on the cross girder from applied loading.
The structural arrangement of the connection between an intermediate cross girder and a main girder is
shown in figure E.12. Only the web of the cross girder is connected to the stiffener on the main girder; the
flanges are not connected.
181
Figure E.12
Main girder and cross girder connection arrangement
dh 26mm
d 24mm
f =
830MPa
uf
2
As = 353m
m
2
Ac = 324m
m
0.35
k h 1.0
Refer to AS 5100.6,
clause 12.5.4.1
Nti = 210kN
SLS slip resistance in double shear:

*
V
Vsf
sf
V
sf
= 0.7
= neiNtiK
V = 0.35 210 1.0 =

73.5kN
sf
Vsf = 51.50kN
Refer to AS 5100.6,
clause 12.5.3.1
E12.2 Shear resistance of bolts

ULS shear resistance of bolts (assuming shear through threads):
*
V
Vf
f
= 0.8
Vf = 0.62fuf kr nnAc + nx Ao
kr =
nn = 1.0
1.0
V = 0.62 830 1.0 1 324 + 0 = 166kN

f
Vf = 133kN
2
r =
n
xn + yn
2
2
r = 616,000m
m
n
rmax
=
702 +2802 = 288.62mm
*
= 2.5% flange compression force distance between the centroid of the top and bottom flanges
M
buckling
182
50 25
10-3 = 117kN.m
= 89
1350
*
M
buckling
*
M
+
shear
= 222
100
140
-3
10
= 38kN.m
*
*
*
M =M
+M
= 117 + 38 = 155kN.m
buckling
shear
155 288.62
3
V
=
10 =
b,m
73kN
616,000
280
Vmx = 73
=
71kN 288.62
70
Vmy = 73
=
18kN 288.62
222
Vs =
=
13.87kN 16
2
2
2
V
= Vm
+
+ Vs
= 71 + 18 +
13.87
Vm
b,max
x
y
= 78kN
V
V
b,max
f
E13 Fatigue assessment

E13.1 Basic loadings for fatigue design
Clause 6.9 of AS 5100.2 gives two basic loads as shown in figures E.13 and E.14
Figure E.13
Modified individual A160 heavy axle load
Figure E.14
Modified individual M1600 moving traffic load without the UDL component
183
The number of fatigue stress cycles to be used for the calculation of the fatigue capacity of the structural
element under consideration should be as follows:
(current number of heavy vehicles per lane per day) x 4 x 104 x (route factor)
Dynamic load allowance is given as follows:
0.4 for the A160 axle load
0.3 for the M1600 axle load.
The route factor for the urban road specified as 0.3.

L is the effective span in metres and is defined as:
for positive bending moments, L is the actual span in which the bending moment is being considered
for reactions, L is the sum of the adjacent spans
for cross-girders, L is twice the longitudinal spacing of the cross-girders.
E13.2 Range of effects due to passage of fatigue vehicle

Through inspection it is clear that the fatigue loading of M1600 is more critical, therefore the fatigue
loading is:
0.70 M1600 without UDL (1+ )
The corresponding total number of cycles is:
n = NHVD
per lanex
2 1
0
L -0.5route factor
route factor 0.3
NHVD 1500
Table E.14
Worst bending effects

Span
M y (kN.m)
Range
Table E.15
2028
Worst shear effects
Span
Fz (kN)
Range
390
E13.3 Assessment of structural steel details

E13.3.1
Design stress range mid-span
For the mid-span the length of the effective span is L 24.50m .

Therefore, the total number of cycles is:
184
n
= 1500 2
4
10
24.
5
-0.5
0.3 = 1.82
10
At mid-span there is negligible stress range in the top flange. The stress range in the bottom flange is:
f* =
2028
67
= 30MPa .
fatigue assessment is required.
As f* = 30MPa
27MPa
The worst detail category that might apply is for the weld between the bottom flange and the web, which,
for a manual continuous weld, is category 100 (AS 5100.6, table 13.5.1-B).
Design value of fatigue strength ffc is:
6
6
n = 1.82 10 5 10
= 1.0
frnc =
100
= frnc
fc
23 106
*
f
f
fc
2 3
= 103MPa
=
100
1.82
- OK
E13.4 Assessment of shear connection

The design value of the stress range in shear studs is given 425MPa times the ratio of the longitudinal
shear load on the stud to the nominal static strength specified in AS 5100.6, clause 6.6.3.
E13.4.1
Shear at supports
The range of vertical shear force at the pier is 390kN.

At the pier, the studs are 19mm diameter, in rows of two at 150mm spacing.
*
uL = 390 0.656 = 256kN / m
A t yc
-1
= 0.
656m
It
uLs = 684kN / m
= 1.0
Therefore,
*
fus =
L
256
0.15
425
uLs
425 = 12MPa
2
684
The fatigue strength of the shear stud is given in AS 5100.6, clause 13.6.3 as follows:
2.08
f =
1
22 8
10
2.08
1
22 8
10
10
= 102MPa
1.82
*
fs ff - OK
Therefore, the number of the shear stud needs to be increased to three.
185
Appendix F: Two span multi-girder bridge

F1
Design statement
For barrier design the annual average daily traffic (AADT) is 15,000 vehicles per day with 20%
commercial vehicles. An offset of 1.2m is assumed. Thus, a G9 modified TL4 Thrie beam is
considered.
The bridge clearance is 6.5m from the soffit to the road surface, therefore the impact load does not
need to be considered.
The bridge is being designed in accordance with following standards:
NZTA (2012) Bridge manual
AS 5100.6 (SA 2004)
NZS 3101 (SNZ 2006).
concrete: C40
The construction load is 0.75kN/m2 and the weight of the temporary formwork is assumed to be
hardened concrete.
Pedestrian live load = 5kPa. This value can be reduced when considered in conjunction with road
traffic loading.
Coefficient of linear thermal expansion = 11.7x10-6 per C.
Characteristic value of shrinkage is
sh 0.0002 .
186
F2
The bridge carries a two-lane single carriageway road over another road. The carriageway has 1m wide
shoulders and has a 2m wide footway on either side. A four-girder arrangement has been chosen and a
deck slab of 250mm has been assumed. The deck overhangs 1.6m outside the centrelines of the outer
girders.
Figure F.1
F3
Two lane bridge structural arrangement
Design basis
The bridge is designed in accordance with the Bridge manual, AS 5100.6 and NZS 3101.
Clifton (2007a).
F3.1
Load combinations
Factors and load combinations of actions are given by the Bridge manual, tables 3.1 and 3.2 for the SLS
and ultimate limit state (ULS) respectively.
F3.2
Factors on strength
187
F3.3
structural steel:
grade 300
concrete:
C40
steel reinforcement: grade 500
fy depends on the product material.
ASI design
320MPa for t 11mm; 300MPa for 11mm <t 17mm; and 280MPa for t
capacity
>17mm.
tables
For plates use:

320MPa for t 8mm; 310MPa for 8mm <t 12mm; 300MPa for 12mm <t 20mm; and 280MPa for
t>20mm. Note that designers may wish to use grade 350 steel for better economy.
'
For concrete, f c 40MPa

For steel reinforcement, fy 500MPa
The modulus of elasticity for both structural steel and steel reinforcement is taken as Es 200GPa
The modulus of elasticity of the concrete is given by NZS 3101, section 5.2.3 as:
1.5
'
E = 3320 f +6900
c
c
2300
Ec =32GPa
This 28-day value is used to determine all short-term effects and resistance, and the modular ratio is
thus:
nS 200 / 32 6.25
E
Ec
=
c,L
1+
32
= 10.50
1+ 2
cc
nL = 200 / 10.50 = 19
The conservative value of
cc is given in RRU 70 as 2.00.
F3.4
F3.4.1
Concrete

Surfaces in contact with the ground = 75mm

Surfaces with a damp proof membrane between the ground and the concrete = 50mm
Elsewhere = 40mm
188
F3.4.2
Steel
For appropriate and cost-effective coating system for structural steelwork New Zealand steelwork
corrosion and coatings guide (El Sarraf and Clifton 2011), should be used in conjunction with AS/NZS
2312 (SNZ 2002) and NZS 3404.1 (SNZ 2009).

Table F.1
Load combinations for the serviceability limit state
Table F.2
Load combinations for the ultimate limit state
189
F4
F4.1
Permanent actions
F4.1.1
generally produces adverse effects and international practice is to base it on nominal thickness + 55%.
However, New Zealand practice differs and reference should be made to the Bridge manual, section 3.4.2.
DL 1.550.1323 4.63kPa
F4.1.2
F4.1.3
A nominal value of 2kN/m is assumed for each steel parapet.

Note that this value will increase if a solid concrete crash barrier is used.
F4.2
Construction loads
For global analysis, a uniform construction load of 0.75kN/m2 is assumed during

casting and the weight of the temporary formwork is assumed to be 0.5kN/m2.
Additionally, wet concrete is assumed to have a density of 1kN/m3 greater than that of
Refer to section
6.1.5 in the main
part of the guide
hardened concrete; for a slab thickness of 250mm this adds 0.25kN/m2.

The total construction load is thus: CN = 0.75 + 0.50 + 0.25 = 1.50kPa
F4.3
Traffic loads
F4.3.1
Road traffic
Refer to section
For the road carried by this bridge, the NZTA specifies the HN-HO-72 traffic
6.4.1 in the main
loading but only HN72 has been considered here for simplicity.
part of the guide
F4.3.2
Pedestrian traffic
Pedestrian traffic is represented by the value in the Bridge manual.
FP 5kPa
This can be reduced to 4.0kPa when applied with traffic loading
F4.3.3
Fatigue load
Clifton (2007a)
is recommended
for fatigue assessment

Refer to section
6.4.2 in the main
part of the guide
190
F4.4
Thermal actions
F4.4.1
For change of length in composite sections, the coefficient of linear thermal expansion is 11.7x10-6 per oC.
This is then used for determining soil pressure on integral bridges.
F4.4.2
The vertical temperature difference given in chapter 3 of the Bridge manual is used and the temperature
difference will be considered to act simultaneously with the overall temperature change.
The effects of vertical temperature gradients are derived for the positive differential temperature
conditions (where the top surface is hotter than the average temperature of the superstructure) and for
the negative temperature differential conditions (where the top surface is colder than the average
temperature of the superstructure).
Note: The negative temperature variation is the same as for bridge type 1 in AS 5100.2, figure 17.3.
Figure F.2
F4.5
Seismic actions
Section 5 of the Bridge manual is used to determine the horizontal force
Refer to section 6.5 in the
generated by the seismic action on the bridge superstructure. These forces are
main part of the guide.
transferred into the supports (piers and abutments), which are designed to
resist this horizontal force (as shown in figure 5.3 of the Bridge manual). Therefore, seismic actions acting on
the superstructure are ignored as they are minimal as long as the load path is available to transfer these actions
into the substructure.
191
F5
Figure F.3
21.7m span girder
12.6m pier girder
21.7m span girder
500 x 25
600 x 50
500 x 25
20
20
20
600x 40
700x 60
600x 40
Top rebar
HD16-150
HD25-150
HD16-150
Bottom rebar
HD16-150
HD25-150
HD16-150
Top flange
Web
Bottom flange
The overall girder depth is 1100mm. Cover to the centroid of the top and bottom reinforcement is 55mm.
This complies with NZS 3101.
Note: The above section sizes are based on an initial few iterations to determine the section properties for
this worked example. For an actual bridge, the designer may wish to undertake further iterations in order
to optimise the cross sections used and gain economy.
F5.1
Figure F.4
Bracing arrangements
The above bracing arrangements are assumed for this example.
F6
Beam cross sections

F6.1
Section properties internal main girders
Section properties are required for global analysis. For section analysis, consider the effective section,
allowing for shear lag.
The equivalent spans for effective width are:
Abutment and mid-span sections:
Lc 0.7L1 0.728
19.6m
Refer to AS 5100.6, clause 6.1.7.
192
Hogging section:
Lc 0.25L1 L2 0.2556 14.0m
At mid-span,
min
eff
L c
19600 ;3700;12250 3000mm

;S;12t min
5
5
14000
;3700;12250
At central pier, beff min
2800mm
5
Refer to
AS 5100.6,
clause 4.4.1
Properties for gross sections at the central pier and in the span are tabulated below.
Figure F.5
Internal main girders effective width
Table F.4

Span girder
Pier girder
Unit
57,200
91,800
mm2
ENA
448
485
mm
Ix
1.2E+10
2.1E+010
mm4
Zx, tf
18E+06
34E+06
mm3
Zx, bf
26E+06
43E+06
mm3
Not compact
Compact
5040
11480
Area
ENA height
Section class
Section bending capacity
Msx
kN.m
Slenderness check for span girder:

e,flange
bf tw
2t f
e,t_ flange
fy,flange
250
500 20 280
2 25 250
e,web d1
tw
10.20
e,web
fy,web
250
1035 300
20
250
Refer to
56.68
AS 5100.6,
section
5.1.2.
193
y,flange
re
e,t _ flange
14
(1100 448 25)
e
0.61
y,web
e,flange
e
e e
10.20
S
flange
10.20
y,flange
0.73
Refer to
14
y,flange
e,web
1035
60 60
e
98
y,web r
e 0.61
AS5100.6,
56.68
98
table 5.1
0.58
e,web flange governs
e
y,web
e e
p
S
e 8
p
Therefore, the bare steel section at span girder is not compact.

Slenderness check for bare steel pier girder:
e,flange
bf tw
2t f
e,b_ flange
fy,flange
250
700 20 280
2 60 250
6.00
e
14
y,flange
re
485 60
e,t_ flange
0.43
990
bottom flange is in compression in the pier girder
e
y,web
60
re
d1 fy,web
e,web
tw
250
990 300
54.22
e,web
20 250
60
139.5
0.43
e e
6.00
S
flange
e 8
p
6.0 0.43
14
e
y,flange
54.22
e,web
0.39
e
y,web 139.5
e,flange
e,web
e
y,flange
e
y,web
flange governs
e e
p
S
Therefore, the bare steel section at pier girder is compact.
Table F.5
Composite cross section (short term) sagging (nS=6.25)

Span girder
Pier girder
Unit
177,200
203,800
mm2
ENA
974
892
mm
Ix
3.6E+10
4.9E+10
mm4
Zx, c
95E+06
107E+06
mm3
Zx, tf
286E+06
236E+06
mm3
Zx, bf
36E+06
55E+06
mm3
Compact
13,220
kN.mm
Area
ENA height
Section class
Plastic bending capacity
Mpl
194
The cross section of the span girder is compact, provided the top flange is restrained by shear connectors
within the spacing limits given in AS 5100.6, clause 6.1.8
Figure F.6
Section moment capacity compression zone entirely within the concrete slab
Refer to
AS 5100.6,
appendix E
fy A
dh
'
0.85fcb
dh ds
M Af
P
y g
28057,200
157mm
0.85 40 3000
ds d
250 157
6
h 57,200 280 779
10 13,220kN.m
2
Note: Uncracked pier girder section properties are needed for the calculation of shear flow.
Note: During the construction of a composite bridge, it is quite likely that a beam will change its section
class, because the addition of the deck slab both prevents local buckling of the top flange and
significantly shifts the neutral axis of the section. Typically, a mid-span section could be compact after
casting the slab but not prior to this. As a consequence, checks at intermediate stages of construction
should be based on the relevant classification for the stage being checked.
Table F.6
Composite cross sections (long term) (n =19)

L
Span girder
Pier girder
Unit
96,575
128,550
mm2
ENA
765
696
mm
Ix
2.6E+10
3.5E+10
mm4
Zx, c
44E+06
54E+06
mm3
Zx, tf
78E+06
87E+06
mm3
Zx, bf
34E+06
51E+06
mm3
Area
ENA height
195
Table F.7
Composite cross sections hogging (cracked)

Span girder
Pier girder
Unit
77,817
110,453
mm2
ENA
654
610
mm
Ix
2.1E+10
2.9E+10
mm4
Section modulus, top rebars
Zx, c
33E+06
42E+06
mm3
Zx, tf
47E+06
60E+06
mm3
Zx, bf
32E+06
48E+06
mm3
Not compact
Area
ENA height
Section class
Slenderness check for composite pier girder:
Refer to
e
e,t_flange = 6.14
y,flange
re =
e,b_flange = 6.00
e,web = 54.22
990
=
y,web
e,t _ flange
e
y,flange
6.14
14
e,web
0.44
y,web
clause 5.1.2
14
610 - 60
e
108
AS 5100.6,
0.56
60
60
re
0.56
e,flange
e
Refer to
AS 5100.6,
table 5.1
web governs
y,flange
e,web
108 0.50
y,web
Plastic neutral axis =1050mm
e e
54.22
S
web
1050
60
105060
1.0
990
= 1.0
990
rp =
rp
111
111 = 30
111 =
111
30
e
4.7rp1
1 4.7
4.71.0
pep =4.7r
1.0 1
p
1
e e
p
S
Therefore, the composite steel section is not compact.
F6.2
Primary effects of temperature difference and shrinkage
F6.2.1
Refer to section 7.3.1
For calculation of primary effects, the short-term modulus of concrete is used.
in the main part of
Ec 32GPa
the guide.
Es 200GPa
For each element of the section, calculate stress as strain times modulus of elasticity, then determine the
force and centre of force for that area. The restraint moment in the inner beam, due to the characteristic
values of temperature difference noted in the Bridge manual, section 3.4.6 is:
Assume 50mm blacktop thickness for this case only:
T (320.2d)o C
d 50mm
T 22o C
The moment release stress is shown diagrammatically below.
196
Figure F.7
The effects of the negative vertical temperature variation are insignificant and therefore have been
ignored.
Table F.8
Temperature induced force and moment effects

Span girder
Pier girder
Axial fixity force
(kN)
4196
4182
Fixity moment
(kN.m)
-1043
-1328
The release of the restraint moments is applied along the span to determine the secondary effects of the
vertical temperature difference.
Note: The following procedure shows how to calculate the secondary moment on the main girders.
197
Figure F.8
Method to calculate main girder secondary moments
The primary effects of differential temperature through the depth of the cross section of
a member are considered in the design. In addition, the secondary effects in continuous
members, due to redistribution of the moments and support reactions caused by the
Refer to
AS 5100.6,
clause 4.9.1.1.
primary effects are also considered.

F6.2.2
Shrinkage
The effects of shrinkage are calculated for the long-term situation, and where the total effects of
shrinkage are advantageous, they are ignored.
The characteristic value of shrinkage is
sh 0.0002 and the long-term modular ratio is used.
For a fully restrained section, the restraint force and moment in the span and pier girders, and inner
beams, due to the characteristic value of shrinkage strain, are given by:
198
Table F.9
Shrinkage induced restraint force and moment

Force
Strain
Centre of force above NA
Moment
kN
mm
kN.m
@ span girder
0.0002
1943
460
894
@ pier girder
529
1028
Hence the primary effects for the pier girder are:

Table F.10
Primary effects for the pier girder
Pier girder
Zx
Restraint
(mm3)
(MPa)
Bending (MPa)
Axial (MPa)
(MPa)
Top of concrete slab
54E+06
0.8
-0.3
87E+06
2.1
-
0.6
0.8
-0.7
Steel top flange
87E+06
2.1
0
12
15
27
Steel bottom flange
51E+06
-20
15
-5
Top reinforcement
58E+06
18
15
33
Bottom reinforcement
76E+06
13.5
15
28.5
Bottom of concrete slab
Figure F.9
Total
Release of restraint
Primary effect stresses over the depth of the pier girder section
The release of restraint moments is applied along the girder to determine the effects of shrinkage.
Figure F.10
Primary and secondary effects of shrinkage
-1465kN.m
Secondary effects
Primary effects
-1135kN.m
-1135kN.m
-1028kN.m
894kN.m
-894kN.m
199
Note: The previously-shown procedure can also be used to calculate the secondary
Refer to AS 5100.6,
clause 4.9.1.2
effect of shrinkage on the main girders.

When the effects of shrinkage modified by creep adversely affect the structure they
are considered in the manner described for differential temperature effects.
Refer to AS 5100.6,
clause 4.9.2.1
The effects of shrinkage and differential temperature are considered at the SLS for
composite beams with sections that are not compact at internal supports.
Account is taken of the longitudinal shear forces arising from shrinkage and differential temperature
effects in the design of all composite beams for the serviceability.
The longitudinal shear forces arising from the effects of shrinkage and differential
temperature are considered in the design of the longitudinal and transverse
Refer to AS5100.6,
clause 4.9.3.2
reinforcement in the concrete slab.
F7
Global analysis
F7.1
Grillage model
Refer to AS 5100,
clauses 6.4.2 and
6.4.3
A grillage model of the superstructure was created, with longitudinal elements for composite girder and
transverse elements for the concrete slab.
For the crack region over the intermediate support (15% of each span) the composite elements were given
cracked properties but transverse elements remained uncracked.
As the bridge is simply supported at end supports, no concrete diaphragms are provided at both abutments.
Figure F.11
Grillage model of the superstructure

Intermediate support line
End support line
End support line
Longitudinal elements
Transverse elements
3500m
370
0m
m
(composite section)
Dummy edge elements
Note: The grillage model utilised a transverse elements comprising a rectangular slab 3500mm by
250mm. Modern computer models automatically generate section properties; however, these should be
checked manually at least for torsional properties to suit the designers assumptions.
F7.2
Construction stages
It is assumed that the deck will be concreted in two stages: first the whole of span 1, followed by the
whole of span 2. The edge beams will be concreted after span 2. Separate analytical models are therefore
provided for:
Stage 1
All steelwork, wet concrete in span 1
Stage 2
Composite structure in span 1 (long-term properties), wet concrete in span 2
Stage 3
Composite structure in both spans (long-term properties)
Stage 4
Composite structure (short-term properties).

200
F7.3
Analysis results
The following results are for design values of actions, ie after application of appropriate load factors on
nominal values of actions.
Note: the following results may not correlate with a slightly different model and hence should not be
viewed as absolute. They do, however, provide an indication of the order of magnitude expected.
F7.3.1
Stage 1
Selfweight of concrete on span 1
Construction load on span 1
Table F.11
Distance from pier (m)
ULS
SLS
M (kN.m)
F (kN)
M (kN.m)
F (kN)
-2580
685
-1960
523
6.3
1083
478
834
365
16.8
3211
-73
2454
-57
28
-488
-373

Figure F.12
Stage 1 loading
Selfweight of steel beams

Span 2
Span 1
Selfweight of concrete + construction loading
F7.3.2
Stage 2
Selfweight of concrete on span 2
Construction load on span 2
Removal of construction load on span 1.
201
Table F.12
ULS
SLS
My (kN.m)
F z (kN)
M y (kN.m)
F z (kN)
-2231
-15
-1714
-12
6.3
-1926
87
-1480
67
16.8
-949
90
-729
69
28
87
67

Figure F.13
Stage 2 loading
Selfweight of concrete + construction
Span 1
F7.3.3
Span 2
Stage 3
Removal of construction loads on span 2.
Figure F.14
Stage 3 loading
Footpath construction
Carriageway
202
Table F.13
ULS
SLS
My (kN.m)
F z (kN)
M y (kN.m)
F z (kN)
-1696
286
-1256
212
6.3
-76
220
-56
163
16.8
1120
-14
829
-10
28
-158
-117
F7.3.4
Long-term shrinkage
The nominal value of secondary moment is calculated using the procedures in figure F.8. These need to be
factored depending on the ULS and SLS load effects
Table F.14
Secondary design actions
F7.3.5
Table F.15
ULS/SLS
My (kN.m)
Fz (kN)
-1465
52
6.3
-1135
52
16.8
-586
52
28
52
Stage 4
HN72 traffic loads for worst hogging at intermediate support
ULS
SLS
My (kN.m)
Fz (kN)
My (kN.m)
Fz (kN)
-3089
754
-1844
451
6.3
1262
648
757
387
16.8
2720
-151
1623
-90
28
-354
-212
Load combination 1A is assumed to be critical
Relative position of vehicle axle loads and pedestrian live load for the worst hogging at intermediate support
203
Table F.16
HN72 traffic load for worst hogging at splice position
ULS
SLS
My (kN.m)
Fz (kN)
My (kN.m)
Fz (kN)
-2052
94
-1226
56
6.3
-1465
88
-873
53
16.8
-684
66
-406
39
28
62
37
Relative position of vehicle axle loads and pedestrian live load for the worst hogging at splice
Table F.17
HN72 traffic load for worst sagging at splice position
ULS
SLS
My (kN.m)
Fz (kN)
My (kN.m)
Fz (kN)
-2032
729
-1215
436
6.3
2104
583
1256
348
16.8
2820
-166
1680
99
-210
28
-352
Relative position of vehicle axle loads and pedestrian live load for the worst sagging at splice
204
Table F.18
HN72 traffic load for worst sagging in span
ULS
SLS
My (kN.m)
Fz (kN)
My (kN.m)
Fz (kN)
-1907
571
-1139
341
6.3
1318
471
785
281
16.8
3798
-217
2262
-130
28
-447
-267
Relative position of vehicle axle loads and pedestrian live load for the worst sagging at mid-span
Table F.19
HN72 traffic load for maximum shear forces
ULS
SLS
My (kN.m)
Fz (kN)
My (kN.m)
Fz (kN)
Pier
-2805
877
-1675
525
Splice/pier side
765
583
458
348
Splice/span side
755
72
454
44
Span
3271
-321
1950
-192
Abutment
-670
-401
Relative position of vehicle axle loads and pedestrian live load for the worst shear at intermediate support
205
Table F.20
Effects of thermal actions
F8
ULS/SLS
My (kN.m)
Fz (kN)
1776
-63
6.3
1376
-63
16.8
710
-63
28
-63
This section provides design values for certain situations in the design of the inner beams. In practice, the
design of other parts of the structure would also need to be considered.
F8.1
Effects of construction loads
Generally, the effects of construction loads apply to different cross-section properties, although for span
1, the cross sections for the inner beam are the same at stages 2 and 3. The following tabulations
summarise the forces and moments at each stage and the stresses due to those effects, for selected cross
sections. According to the Bridge manual the load combination 5B is the most critical load case for
construction stages.
Table F.21
Stress at pier (using steel and cracked concrete section properties)

Bottom flange
My
Fz
Zx, bf
kN.m
kN
Stage 1
-2580
Stage 2
-2231
Top flange
Zx, tf
10 mm
MPa
685
43
-15
Top rebars
Zx, c
10 mm
MPa
10 mm
MPa
-60
34
76
48
60
37
42
53
60
28
42
40
-1696
286
48
46
-35
Shrinkage - PE
-7
36
45
Shrinkage - SE
-1978
70
48
60
33
42
47
Stage 3
41
(not adverse)
Temperature
-8485
1026
-189
185
210
Table F.22
Stress at splice (using steel and cracked concrete section properties)

Bottom flange
Stage 1
Stage 2
Stage 3
Shrinkage - PE
Top flange
Top rebars
My
Fz
Zx, bf
Zx, tf
Zx, c
kN.m
kN
106mm3
MPa
106mm3
MPa
106mm3
MPa
1083
478
43
25
34
42
46
-1926
76
-
87
220
-
48
60
32
32
48
40
-
60
42
2
-7
36
45
206
Bottom flange
Shrinkage - SE
-1532
70
48
Top flange
60
Top rebars
26
42
36
32
(not adverse)
Temperature
-2451
855
-56
63
129

Table F.23
Stress at span girder (using steel and long-term concrete section properties)
Bottom flange
Stage 1
Stage 2
Stage 3
My
Fz
Zx, bf
kN.m
kN
3211
-949
1120
Zx, tf
10 mm
MPa
26
73
90
-
Temp - PE
Temp - SE
316
3698
-28
Top rebars
Zx, c
10 mm
MPa
10 mm
MPa
124
18
-178
34
78
12
44
22
34
28
33
78
-14
44
14
Shrinkage
Top flange
(not adverse)
34
4
9
78
4
-
44
4
-186
142
25
1
7
9
25be critical
Load combination 5B is assumed to
As the support at each abutment is assumed to be simply supported, the stress at the abutment will be
zero.
F8.2
Effects of traffic load plus construction stages loads (ULS)
The worst effects are due to HN72 traffic loads.

Table F.24
Loading for maximum hogging at pier (using cracked concrete section properties)
Bottom flange
Top flange
Top rebars
My
Fz
Zx, bf
Zx, tf
Zx, c
kN.m
kN
106mm3
MPa
106mm3
MPa
106mm3
MPa
Stages 1, 2, 3
-6507
956
-141
141
93
HN72 traffic
-3089
754
48
60
51
42
74
64
-7
36
45
60
33
42
47
Shrinkage - PE
Shrinkage - SE
-1978
70
48
-11,574
1780
41
-254
219
259
207
The maximum hogging moment at the splice position is much greater than the maximum sagging
moment, so this will govern the design of the splice.
Table F.25
Effects at splice position (using cracked concrete section properties)

Bottom flange
Stages 1, 2, 3
HN72 traffic
Top flange
Top rebars
My
Fz
Zx, bf
Zx, tf
Zx, c
kN.m
kN
106mm3
MPa
106mm3
MPa
106mm3
MPa
-919
785
-17
48
42
35
-1465
88
48
60
1
24
31
-7
36
45
60
26
42
36
Shrinkage - PE
Shrinkage - SE
-1532
70
48
-3916
32
-86
943
164
88

Table F.26
Loading for maximum sagging bending (using short-term composite section properties)
Bottom flange
Top flange
Top rebars
My
Fz
Zx, bf
Zx, tf
Zx, c
kN.m
kN
106mm3
MPa
106mm3
MPa
106mm3
MPa
Stages 1, 2, 3
3382
129
-180
-4
HN72 traffic
3798
-217
36
106
286
95
-40
Shrinkage
(not adverse)
7180
-214
235
13
-193
-44
F8.2.1
Loading for maximum shear
The value of maximum shear is needed to verify the shear resistance of the web and to determine the
longitudinal shear on the stud connectors.
Table F.27
Maximum shear at pier position

My
Fz
kN.m
kN
Stages 1, 2, 3
-6057
956
HN72 traffic
-2805
877
Shrinkage
-1978
70
-10,840
1903
208
The value of the maximum shear is needed to determine the maximum longitudinal shear on the stud
connector.
Table F.28
Maximum shear at splice position

My
Fz
kN.m
kN
Stages 1, 2, 3
-919
785
HN72 traffic
765
583
-1532
70
-1686
1438
Shrinkage
The value of the maximum shear is needed to determine the maximum longitudinal shear on the stud
connector.
Table F.29
Maximum shear at span girder

My
Fz
kN.m
kN
Stages 1, 2, 3
3382
HN72 traffic
3271
-321
Shrinkage
(not adverse)
6653
-318
Note; Load combination 1A is assumed to be critical

Table F.30
Maximum shear at abutment

My
Fz
kN.m
kN
Stages 1, 2, 3
-559
HN72 traffic
-670
Shrinkage
(not adverse)
0
-1229
F8.3
Effects of traffic load plus construction stages loads (SLS)
The values of effects at SLS are needed to verify crack control in the slab at the pier and to verify the slip
resistance of the splice.
209
Table F.31
Effects at pier position (worst shear)

Bottom flange
My
Fz
Zx, bf
kN.m
kN
Stage 1
-1960
Stage 2
Stage 3
Top flange
Zx, tf
10 mm
MPa
523
43
-1714
-12
-1256
Shrinkage -PE
Shrinkage - SE
HN72 traffic
Top rebars
Zx, c
10 mm
MPa
10 mm
MPa
-46
34
58
48
60
29
42
41
212
48
36
-26
60
21
42
30
-5
27
33
-1465
52
48
60
24
42
35
48
31
-38
60
31
42
44
-1844
525
-8239
1300
-182
183
190
Note; Load combination 1A is assumed to be critical

Table F.32
Effects at splice position (worst shear)

Bottom flange
My
Fz
Zx, bf
kN.m
kN
834
365
Stage 1
Stage 2
-1480
Stage 3
67
163
Top flange
Zx, tf
10 mm
MPa
10 mm
43
19
Zx, c
MPa
10 mm
MPa
34
42
35
60
48
31
-
60
42
1
-5
27
33
19
42
27
42
-11
Shrinkage - SE
-1135
52
48
60
48
24
10
60
-1379
348
48
Shrinkage -PE
458
25
25
56
-
HN72 traffic
Top rebars
995
-32
8
39
85
F9
Verification of bare steel girder during

construction
See figure 7.4 in the

The paired beams are susceptible to lateral torsional buckling under the weight of the wet concrete (ie
before it hardens and provides restraint to the top flanges).
The beams are partially restrained against buckling by the bracing frames between each pairs at three
points in the span.
Use gross section properties for verification of buckling resistance.
F9.1
Evaluation of lateral restraint
If the length of the segment between braces satisfies the requirements in
Refer to section
7.1.2 in the main
part of the guide.
AS 5100.6, clause 5.3.2.4, the nominal member moment capacity of a segment is

taken as the nominal section moment capacity otherwise the member moment capacity is calculated using
the expression in AS 5100.6, clause 5.6.1
210
(80 + 50 ) 2pdfA 250
ry
2.5Zex
if the segment is of unequal flanged I section
clause 5.3.2.4
fy
this value can be conservatively taken as -1.0
m 0.2
Refer to AS 5100.6,
Refer to AS 5100.6,
Ls,max 7367mm
figure 4.2.2.2
fy 280MPa
ry
Iy
A
9.81E 08
57200
131mm
Ls ,max
ry
7367
56
131
25500
12
9.8110 8
0.26
df 1067.5mm
2 0.66 1067.5 57,200 250

250
(80 + 50m)2pdfA
= 60
= 90
L
ry
2.5Zex
fy
(80 + 50 )
F9.2
250
2.5 18 106
M M
fy
Verification
Ze,x min Zx,tf ; Zx,bf

6
3
Z
1810 m
m
x,tf
6
3
2610 m
Z
m
x,bf
6
3
1810 m
Z
m
e,x
Ms Ze,x fy 5040kN.m
280
8
4
Iy 9.8110 mm
Iw 0
3
4
J 1224010 mm
Mb Ms 5040kN.m
E = 200,000MPa
G = 80,000MPa
M = 3211kN. m < Mb = 5040kN. m OK
F10 Verification of composite girder

See figure 7.5 in the
F10.1 In hogging bending
The composite section in hogging is not compact.

the stages.
From sections F8.1 and F8.2, the design moment on the steel section only is 2580kN.m and the total
hogging moment is 11,574kN.m, which means that the moment on the composite (cracked) section is
8994kN.m. The stresses are shown in figure F.15:
211
Figure F. 15
Bare steel and composite stresses in hogging
Refer to
AS 5100.6,
clause 6.2.2.
As the composite section in hogging region is not compact, for verification of cross-section resistance, the
stresses should not exceed the limiting stress.
fs
fs
M
fb b fy
Zenc
y
b
Z ec
Mb Mds Z es
Ms
Refer to
AS 5100.6,
clause 6.4.2.2.
Zes min(1.5Zx,tf ;1.5Zx,bf ; Sx )

Zes min(1.534E 06;1.5 43E 06; 41E 06) 41E
06
Zec 48E 06
Mss
Mds = dMss Mss

1
Lb 2 dw 3
d 0.018
r t 0.4
y w
ry
60
3
700
12
50
700
28, 000 2 990 3

d 0.018

0.4 0.38
202 20
d 0.8
202mm
6
4110 280
11, 480kN.m
6
10
d 0.8
4
2
d 3 d
4
2
0.38 3 0.38
1.27
d 1.00 d 1.00
Mds Mss 11, 480kN.m
Ms fyZenc
Zenc 48E 06
6
280 48 10
Ms
13, 440kN.m
6
10
48
13, 440kN.m Ms 13, 440kN.m OK
41
Mb 11, 480
13, 440 10
fb
48
6
10
6
280MPa fy
Therefore, in the steel section:
Refer to
AS 5100.6,
clause 6.3.4
212
fs,bf
= 254MPa f = 0.9 280 = 252MPa OK

b
= 219MPa < f = 0.9 280 = 252MPa OK
fs,tf
And in the reinforcement
= 259MPa f = 0.9 500 = 450MPa OK
fs,reo
Refer to section
F10.2 Maximum shear at support
7.4.1 in the main
The maximum shear in the girder at the intermediate support is 1903kN.
part of the guide.
The thickness of the unstiffened web shall satisfy the requirement in AS 5100.6, clause 5.9.1.
There are no longitudinal stiffeners intended to be used.
d f
t w 1 y
180 250
Refer to
tw 20mm
AS 5100.6,
clause 5.9.1.
dp fy 990 300
6.0mm
250 180 250

180
dp fy
6.0mm OK
tw 20mm
250
180
dp
82
Vw if
tw
fy
250
Vv Vu
V if dp
82
b
fy
tw
250
dp
tw
82
990
49.50
20
fy
250
82
74.86
300
250
Vu Vw
Vw 0.6fy A w
0.6 99020 300
Vu Vw
3564kN
1000
V = 1903kN Vu = 0.9 3564 = 3208kN OK
213
F10.3 Combined bending moment and shear

When the maximum moment is assumed to be resisted by the whole of the cross
Refer to section
7.4.2 in the main
part of the guide.
section, the member will be designed for combined bending and shear, and satisfy the requirement in AS
5100.6, clause 5.11.3.

F10.3.1
Maximum shear with coexisting moment
*
M 10, 840kN.m
V 1903kN
Ms 0.9 13, 440 12,
096kN.m
0.75Ms 9072kN.m
M = 11,123kN. m 0.75Ms = 9072kN. m
Refer to AS 5100.6,
Therefore,
clause 5.11.3.
Vvm
Vvm Vv 2.2
1.6M *
Ms
Vv 3564kN
1.6 10, 840
12, 096
2.
Vvm 3564 2
2731kN
Vvm 2458kN
V Vvm OK
Therefore, the shear capacity of the web is sufficient.

In order to increase the shear capacity, a transverse web stiffener is provided at 1967mm from the
support (ie divide the length to the first bracing into three panels).
An intermediate web stiffener has an area that satisfies the requirement in AS 5100.6, clause 5.14.3.
As 0.5yAw(1 v )
*
Vu dp
s 1967mm
1 S
d
p
dp 990mm
0.75
82
v
2 1.01.0

dp
s
fy
tw 250
dp
v 1.0
1.98
dp

82
0.75
1.0 1.0
v 990 300
2
1967
250
20
990
2
As 0.0mm
Also, an intermediate web stiffener should satisfy the requirement in AS 5100.6, clause 5.14.5 for the
minimum second moment of area about the centreline of the web.
214
1.98 2 1.41
d1
33
5d1tw
2
s
3
1.5 990
6
4
3 10 mm
3
20
2
1967
1.
Is
Is
Try a 250mm x 16mm stiffener plate at each side of the web.

2
16 2503
6
250 16
Is 2
183.20 10
16 250

mm
12
6
3 10
mm
Refer to AS 5100.6,
4
OK
clause 5.14.5
Therefore, the shear capacity of the stiffened web is:
Vv Rsb Vw
The effective length of the web is taken as:
17.5t s
17.5 20 1967
w
; min
le min
;
320mm
fy 2
2
300
250
250
The strut effective cross section, As, is taken as the stiffener area plus the effective length of the web.
A 2 250 16320 2014400m
m
s
Therefore,
Rsb cAsfy
b 0.5
6
I 183.20 10 m
m
s
L e
Is
183.20
6
10
As
14400
rs
113mm
rs
2100
a =
k f 1.0
fy 320
300
Refer to AS 5100.6,
kf
1.0 3.10
250
250
113
clause 5.13.2.
2100(3.10 13.5)
13.5
=
= 10.85
2 15.3 + 2050 3.102 15.3 3.10 + 2050
ab 3.10 0.5 10.85 2.33
0.00326 13.50
0
+1+
2
90
2 2
90
2.33
+1
2
90
2.33
= 746.5
2
2
0.9 14, 400 300

R
3888kN
sb
90
2
746.5 1 1
1.0

746.5 (2.33)
c 1 1
90
90
1000
Vv 3888 3208 7096kN
215
Please note that the web connection of intermediate transverse stiffeners should be designed to resist a
design shear force per unit length (kN/mm) of not less than:
2
0.0008 twfy
bes
16
bes 250
242mm
2
2
0.0008 20 300
w
0.396kN / mm
242
F10.3.2
Maximum moment with coexisting shear
*
V 1780kN
*
M 11, 574kN.m
Ms 0.9 13, 440 12,

096kN.m
0.75Ms 9072kN.m
M = 11407kN. m 0.75Ms = 9072kN. m
Refer to AS 5100.6,
clause 5.14.8
Therefore,
V Vvm
Vvm Vv 2.2
1.6M *
Ms
2.
Vvm 3564 2
Vvm 2146kN
1.6 11574
12096
2384kN
V Vvm OK
F10.4 Sagging bending moment

Refer to AS 5100.6,
The PNA of the composite cross section in sagging is entirely within the concrete
clause 5.11.3.
slab so, the plastic resistance can be utilised.

The short term plastic resistance of the composite section is 0.9 x 13,220=11,898kN.m and the total
design value of bending effects is 7180kN.m, with a very small shear force, so the
section is satisfactory on inspection.
It can be seen that the stresses calculated elastically, taking account of
construction in stages, are also satisfactory as shown in figure F.16.

of the main part of
the guide.
216
Figure F.16
Sagging bare steel and composite stresses
F10.5 Verification of crack control at SLS

F10.5.1
Minimum reinforcement spacing
The spacing of reinforcement crossing a potential crack and located next to the tension face should be
smaller than the values given in NZS 3101, clause 2.4.4.4.
fs 183MPa (max SLS stress at top rebars)
cc 55
90, 000
70, 000
s min
2.5cc ;
fs
fs
25
2
42.5mm
cm 55mm
90, 000
s min
70, 000
2.542.5;
183
183 382mm
sprovided 150mm srequired 382mm OK

F10.5.2
'

Crack control
y kd
d kd

in the main part of the
1350 1350 696
1.22
1225 1350 696
guide.
kd depth of NA for long term composite section

2
s
2
150
2
g s cm
55 93
2
2
w = 2.0
fs
g
E
s
= 2.0 1.22
183
200,000
93 = 0.21mm
w 0.21mm 0.3mm OK
F11 Longitudinal shear
Refer to sections 6.1.7 and
guide.
7.7 in the main part of the
shear connectors and the transverse reinforcement at the pier, at the splice and at mid-span.
217
F11.1 Shear forces

Table F.33
ULS values
Pier
Splice
Span
Abutment
685
478
-73
341
377
76
488
-
877
583
-321
71
-
1903
1438
318
670
-1229
Pier
Splice
Span
Abutment
523
365
-57
Table F.34
SLS values

252
282
59
373
-
525
348
-190
50
-
1300
995
-188
401
824
F11.2 Section properties

To determine shear flows the parameter
At yc It is needed for each section and stage.
For composite sections, uncracked unreinforced composite section properties can be used to determine
shear flow.
Table F.35
At yc It for each section

Span girder/abutment
Pier girder
girder
Web/top fl
Top fl/slab
Web/top fl
Top fl/slab
At yc It
At yc It
At yc It
At yc It
m-1
m-1
m-1
m-1
Steel section
0.845
0.689
Long-term section
0.882
0.557
0.854
0.698
Short-term section
0.873
0.761
0.876
0.837
F11.3 Shear flow at ULS

Table F.36
Force at web/top flange junction
At pier
685 x 0.845 + 341 x 0.882 + 877 x 0.873 =
1645
kN/m
At splice
478 x 0.689 + 377 x 0.854 + 583 x 0.876 =
1162
kN/m
At mid span
-73 x 0.689 + 76 x 0.854 321 x 0.876 =
-267
kN/m
At abutment
-488 x 0.689 -71 x 0.854 670 x 0.876 =
-984
kN/m
218
Table F.37
At pier
341 x 0.557 + 877 x 0.761 =
At splice
857
kN/m
377 x 0.698 + 583 x 0.837 =
751
kN/m
At mid span
76 x 0.698 321 x 0.837 =
-216
kN/m
At abutment
-71 x 0.698 670 x 0.837 =
-610
kN/m
At pier
252 x 0.557 + 525 x 0.761 =
540
kN/m
At splice
282 x 0.698 + 348 x 0.837 =
488
kN/m
At mid span
59 x 0.698 192 x 0.837 =
-120
kN/m
At abutment
-50 x 0.698 401 x 0.837 =
-371
kN/m
F11.4 Shear flow at SLS

Table F.38
F11.5 Web/flange weld

For 6mm throat SP fillet weld tw=8mm
w
6mm
1.11kN / mm
Resistance of two welds = 2220kN/m >1645kN/m, shear flow in pier girder at top flange OK
By inspection, 5mm throat SP fillet weld would be satisfactory at the splices and in the span regions.
F11.6 Shear connectors
Refer to section
7.7.3 in the main
Shear stud connectors 19mm diameter, 150mm long are assumed, with fu=410MPa.
part of the guide
2
fks 0.63 d s fuc
b
or
'
fcy 40MPa
d2 f ' E
fks 0.63 bs cy c
fu 410MPa
c
dbs 16mm
'
Ec 5050 fcy 5050 40 32, 000MPa
Note: for the purpose of evaluating the resistance of headed stud connectors embedded in solid slabs, the
modulus of elasticity for concrete may be taken to be E c
5050 fcy' . From structural reliability analyses
conducted by HERA, it has been shown that the use of this value produces predictions that better reflect
the resistance of studs in physical tests.
Therefore the design resistance of a single headed shear connector is:
0.63 192 410; 0.63

fks min
2
19
40 32, 000
93.25kN
Refer to AS 5100.6,
clause 6.6.3.2.
Longitudinal shear
forces due to the primary effects of shrinkage or differential temperature are

assumed to be transmitted across the interface between the steel
Refer to AS 5100.6,
clause 4.9.2.2
219
beam and the concrete slab by shear connectors at each end of the beam, ignoring the effects of bond.
In the absence of a more accurate analysis, the forces on the connectors may be calculated by assuming
that the rate of transfer of longitudinal force varies linearly from a maximum at the end of the beam to
zero at a distance from the end equal to the total effective width of the slab. Alternatively, where stud
shear connectors are used, the rate of transfer of force may be assumed to be constant over a distance
from each end of the beam equal to one fifth of the span of the beam.
In this exercise, it is assumed that longitudinal shear flow is constant over the span of the beam.
*
L Ls
Ls 0.55nfks
1.0
1943 5
kN
/
m
uL = 540kN / m +
Shrinkage PE
282
*
= 713kN /
(load combination 1A is assumed to be critical)
The shear connectors between the point of maximum negative bending moment and the adjacent point of
zero moment, at the SLS, shall satisfy the requirement in AS 5100.6, clause 6.6.3.3:
For simplicity, it has been conservatively assumed that the negative moment envelop is within the cracked
section regions.
*
Fh = 0.55Arsfsy
2
m
Ars = 2 3272 =
6544m
m
*
-3
F = 0.55 6544 2.8 500
= 5039kN
10
h
fsy = 500MPa
*
f = 5039 (2 0.15 28) = 600kN m
h
then a row of three studs has a design resistance of:
0.55 93.25 3
uLs =
= 1026kN /
m 0.15
*
uL = 713kN / m u = 1026kN / m - OK
Shear
connectors have been designed for critical location; however, further economies can be realised
L
through adjusting the spacing to suit demand along the length of the girder.
F11.7 Transverse reinforcement

Consider the transverse reinforcement required to transfer the full shear
resistance of the studs, ie 1026kN/m as well as the maximum ULS shear flow, ie
1028kN/m.

in the main part of the
guide.
220
Figure F.17
Transverse reinforcement critical shear planes
AS 5100.6, clause 6.6.5.2 as follows:
*
uLp 0.9us + 0.7Atsfry
and
'
*
uLp 0.15uf c
Refer to AS 5100.6,
clause 6.6.5.2
Assumed HD20-150mm:
s = 1MPa
2
u = 2 150 + 400 = 700mm
/m
A = 2 314 150 =
4188m
m
st
fry 450MPa
0.7 4188 450
u = min 1.0 0.9 700 1.0 +

L

3
10
;1.0 0.15 700 40= 1949kN /m
1.35 1943 5
*
uLp = 857kN / m +
= 1091kN / m
Shrinkage PE
28 2
*
= 1091kN / m u = 1949kN / m - OK
uLp
L
F12 Main girder splices
Refer to section
7.8.4 in the main
F12.1 Forces and moments at splice position

Worst hogging moment at splice, at ULS and SLS
Worst shear at splice, at ULS and SLS
The worst sagging moment is much less than maximum hogging moment.
The in-service design combination of actions considered is Group 1A loading:
U1A = 1.35 DL + 1.67LL I + SG + 1.30FP

S1A = DL + 1.35LL I + SG + FP
part of the guide.
221
Table F.39
Main girder stresses and shears at splice location

ULS hog
SLS hog
ULS shear
SLS shear
88
62
76
55
MPa
Top flange stress

Bottom flange stress
Shear force
-32
MPa
86
943
59
700
40
1439
995
kN
From the above stresses, the forces in each flange and moment in the web are as follows:
Table F.40
Main girder forces and moments at splice location

ULS hog
SLS hog
ULS shear
SLS shear
Top flange force
2640
1860
2280
1650
kN
Bottom flange force
-3612
-2478
-1680
-1344
kN
Web force
Web moment
kN
36
256
41
179
367
235
236
163
kN.m
It is noted that the compressive stress in the top flange is higher at construction stage 1, under wet
concrete load in span 1. At this stage, the splice must provide continuity of stiffness, without slipping, and
because the beams are slender at that stage it is appropriate to amplify the design force to ensure
adequate continuity of resistance.
The maximum ULS stress in the top flange during construction is 32MPa. Thus the ULS design force for
the top flange is 960kN.
The maximum ULS stress in the bottom flange during construction is 25MPa. Thus the ULS design force
for the bottom flange is 1050kN.
The maximum SLS stress in the top flange during construction is 25MPa. Thus the SLS design force for the
top flange is 750kN.
The maximum SLS stress in the bottom flange during construction is 19MPa. Thus the SLS design force for
the bottom flange is 798kN.
F12.2 Slip resistance of bolts

Use M24 grade 8.8 tension-friction bolts in double shear in normal clearance holes.
d 24mm
d =
26mm h
f = 830MPa
uf
2
A = 353m
m
s
2
A = 324m
m
c
= 0.35
assumed clean, as rolled surface
k = 1.0
h
Nti = 210kN
SLS slip resistance in double shear:
Refer to AS 5100.6,
V
sf
sf
clause 12.5.4.1
= 0.7
222
V = neiNtiK
sf
h
V = 0.35 2 210 1.0 = 147kN

sf
V = 103kN
sf
F12.3 Shear resistance of bolts
Refer to AS 5100.6,
clause 12.5.3.1
ULS shear resistance of bolts (assuming shear through threads)

*
V
V
f
0.8
Vf 0.62fufkr nnAc nx Ao
kr 1.0
nn 2.0
Vf 0.628301.02324 0 333kN
Vf 266kN
F12.4 Bolt spacing and edge distances

End and edge distance: 1.75df = 42mm for sheared or hand flame cut edge
Spacing:
Refer to AS 5100.6,
clause 12.5.2.2 and
AS 5100.6, table 12.5.2.
2.5df 60mm
Note: spacing of 70mm is utilised to reflect New Zealand industry standard gauge lines.
F12.5 Splice configuration

Consider the following splice configuration
Figure F.19
F12.5.1
Splice configuration
Top flange splice
Dimension for lower covers

Bolt spacing:
Refer to AS 5100.6,
clause 6.6.2.
223
e1 50mm
e2 60mm
p1 70mm
p2 75mm
Overall dimension = 600 x 195 mm

Thickness = 10mm
The length of the cover plate is sufficiently short such that shear connectors do not need to be welded to
the cover plate. The maximum permitted longitudinal spacing is 600mm
Note: The tensile capacity of the cover plate also needs to be checked.
F12.5.2
Bottom flange splice
Dimension for upper covers
Bolt spacing:
e1 50mm
e2 60mm
p1 70mm
p2 75mm
Overall dimension = 860 x 195 mm

Thickness = 20mm
F12.5.3
Web splice
Refer to AS 5100.6,
Bolt spacing:
e1 50mm
clause 6.6.2(g)
e2 50mm
p2 75mm
Overall dimension = 210 x 925mm
Thickness = 10mm
F12.6 Verification of connection resistances

F12.6.1
Top flange splice
There are four rows of bolts, with four bolts per row across the flange.
Slip resistance at SLS = 20 x 103 = 2060kN >1860kN OK
Resistance at ULS = 20 x 266 = 5320kN >2640kN OK
F12.6.2
Bottom flange splice
There are six rows of bolts, with four bolts per row across the flange
Slip resistance at SLS = 24 x 103 = 2472kN
2478kN OK
Resistance at ULS = 24 x 266 = 6384kN >3612kN OK

F12.6.3
Web splice
The splice has a single column of 12 bolts at 75mm spacing.
max
For this group the modulus for the outer bolts =

Where
r2 r
ri is the distance of each bolt from the centre of the group and rmax is the distance of the
furthest bolt from the centre of the group.

224
Here, the modulus = 1950mm.

The extra moment due to the shear = shear force x eccentricity of group from the centreline of the splice.
Hence the force on the outer bolts at ULS and SLS are as shown in table F.41.
Table F.41
Web splice design actions

ULS hog
SLS hog
ULS shear
SLS shear
943
700
1439
995
kN
-36
-41
-367
-236
kN
256
179
235
163
kN.m
52
39
79
55
kN.m
Total moment M
308
218
314
218
kN.m
Force per bolt due to M
158
112
161
112
kN
31
20
kN
161
116
192
132
79
58
120
83
kN
(= V / 12)
179
130
226
156
kN
(Vector sum)
Shear V
Longitudinal force F
Moment
Moment due to e = 55mm
Force per bolt due to F
Total horizontal force H

Vertical force due to V
Resultant force
(= M / 1950)
(= F / 12)
L
By inspection the web splice is satisfactory.
F13 Bracing
Refer to sections 7.3.5, 7.3.6 and 7.3.7 in

the main part of the guide.
The configuration of the intermediate bracing system between girder pairs is as shown in figure F.20.
Figure F.20
Bracing system between girder pairs
Assume the use of 125 x 125 x 10 equal angle sections.

Here the bracing is only used to provide stability under the construction load.
*
Mstage1
3211kN.m
Ztop _ flange 18106mm3

Zbot _ flange 26106mm3
Stress at the top/bottom flange is:
top _ flange 178MPa

bot _ flange 124MPa
Therefore the maximum flange force is:
225
Refer to AS 5100.6,
178 500 25
F =
=
2225kN
f
1000
clause 8.4.2
2.5%Ff 56kN
ASI design capacity
fNc = 129kN
tables
The buckling resistance of the 3900mm diagonal is easily adequate for this force and two-bolt end
connection will also be adequate.
Refer to sections
F14 Fatigue assessment
6.4.2 and 7.11.1 in

the main part of the
F14.1 Basic loadings for fatigue design
guide.
Clause 6.9 of AS 5100.2 gives two basic loads as shown in figures F.21 and F.22.
Figure F.21
Modified individual A160 heavy axle load
160 kN
Elevation
1800 mm
400 mm
3000 mm standard
design lane
250 mm
Plan
Figure F.22
Modified individual M1600 moving traffic load without the UDL component
360 kN
360 kN
360 kN
360 kN
Elevation
3750
12501250
Varies 6250 min.
12501250
5000
1800 m m
400 mm
12501250
12501250
3000 mm standard
design lane
250 mm
Plan
The number of fatigue stress cycles to be used for the calculation of the fatigue capacity of the structural
(current number of heavy vehicles per lane per day) x 4 x 104 x (route factor).
Dynamic load allowance is given as following:
226

For positive bending moments, L is the actual span in which the bending moment is being considered.
For negative moment over interior supports, L is the average of the adjacent spans.
For end shear, L is the actual span.
For reactions, L is the sum of the adjacent spans.
For cross-girders, L is twice the longitudinal spacing of the cross-girders.
F14.2 Range of effects due to passage of fatigue vehicle

By inspection it is clear that the fatigue loading of M1600 is more critical, therefore the fatigue loading is:
0.70 (M1600 without UDL) (1 + )

The corresponding total number of cycles is:
n = NHVDper lanex 2104 L0.5 route factor

NHVD 1500
Table F.42
Lane 1
Lane 2
Table F.43
Lane 1
Lane 2
route factor 0.3

Pier
Splice
Span
M (kN.m)
M (kN.m)
M (kN.m)
Positive
403
821
Negative
-1035
-442
-190
Range
1035
845
1011
Positive
361
822
Negative
-392
Range
905
905
753
162
984
Pier
Splice
Span
F (kN)
F (kN)
F (kN)
Positive
234
144
63
Negative
26
-75
Range
234
118
138
Positive
194
135
58
Negative
0.8
24
111
53
111
Worst shear effects
Range
193
227
F14.3 Assessment of structural steel details

Design stress range at pier:
For the intermediate support the length of the effective span is L 28.0m .
n =1500 2104 28-0.5 0.3=1.7 106
At the pier, the stress range f*in top and bottom flanges (at their mid thickness) is:
f* =
Top flange:
f* =
Bottom flange:
1035
60
1035
48
=17MPa
= 22MPa
As f* 22MPa 27MPa , fatigue assessment is not required; however, the detailed calculation is included
for reference.
The worst detail category that might apply is for a bearing plate welded to the underside of the bottom
flange, which, for a flange plate over 25mm thick, is category 36 (AS 5100.6, table 13.5.1-B).
Design value of fatigue strength
ffc is:
Refer to AS 5100.6,
frn = 36MPa
clause 13.1.7.
0.25

25
=
B
tf
0.25
25
t
f
= 0.80
60
frnc = Btf frn = 0.80 36 = 29MPa

6
6
n = 1.7 10 5 10
= 1.0
fc = frnc
2
6 3
=
29
10
= 31MPa
1.7
frn f - OK
fc
Refer to AS 5100.6,
F14.4 Design stress ranges at splice position
clause 13.7.2
At the splice position, there is negligible stress range in the top flange. The stress range in the bottom
flange is f* =
845
36
= 24MPa . Again as the design stress range is less than 27MPa, the fatigue assessment is
not required. However, for information, the detailed calculation is included as follows:
The most onerous detail at a bolted splice would be category 140. Thus the fatigue strength is:
f
fc
2 6 13
10
rnc
n
= f
-0.5
n =1500 210 28
0.3 =1.7 10
6
6
n =1.7 10 510
frn =140MPa
228
0.25 0.25
25
25
= 0.84
tf
t
f
50
Refer to section
frnc = tffrn = 0.84 140 = 117.6MPa
7.7.6 in the main
1.0
2 6 13
2 13
=177MPa
f = frnc 10 =117.6
1.7
fc
n
part of the guide.
*
f n f - OK
fc
F14.5 Design stress ranges at mid span

At mid-span there is negligible stress range in the top flange. The stress range in the bottom flange is
f* =
1011
36
= 28MPa .
The most onerous detail would be a transverse web stiffener, for which the fatigue category is 71. Thus
the fatigue strength is:
f
= frnc
fc
10
n = 1500 2 10
6 13
28-0.5 0.3 = 1.7
10
n 1.710 510
6
frn 71MPa
0.25
0.25
25
25
0.84

tf
50
f
frnc = tffm = 0.84 71 = 60MPa
1.0
2106 13
2 13

ffc frnc
60
63MPa
n
1.7
frn ffc OK
F14.6 Assessment of shear connection

The design value of the stress range in shear studs is 425MPa times the ratio of the longitudinal shear
load on the stud to the nominal static strength specified in AS 5100.6, clause 6.6.3.
F14.6.1
Shear at pier
The range of vertical shear force at the pier is 234kN.

At the pier, the studs are 19mm diameter, in rows of three at 150mm spacing.
* 2340.761 178kN / m
L
A
y
t c
0.761m1
It pier girder, short-term

229
Ls 1026kN / m
1.0
Therefore,
*
L
fs* 425
Ls
1780.15
3
425 4MPa
1026
The fatigue strength of the shear stud is given in AS 5100.6, clause 13.6.3 as follows:
18
22 18
2.081022
2.08106
ff
1.7 10 103MPa
n
f f OK
F14.6.2
Shear at splice
As the shear force at splice position is lower than that calculated above, it is apparent that the fatigue is
not a matter of concern in this position.
F14.6.3
Shear at mid-span
As the shear force at mid-span is lower than that calculated above, it is apparent that the fatigue is not a
matter of concern in this position.
230
Appendix G: Single span multi-girder bridge

G1
Design statement
For barrier design the AADT is 15,000 vehicles per day with 20% commercial vehicles. An offset of
1.2m is assumed. Thus, a G9 modified TL4 Thrie beam is considered.
The bridge clearance is minimum 1.5m from the soffit to design flood level, therefore flood load does
not need to be considered.
The bridge is being designed in accordance with following standards:
NZTA (2012) Bridge manual
AS 5100.6 (SA 2004)
NZS 3101 (SNZ 2006).
concrete: C40
Construction load of 0.75kN/m2 and the weight of the temporary formwork is assumed to be
hardened concrete.
Pedestrian live load = 5kPa
This value can be reduced when considered in conjunction with road traffic loading.
Coefficient of linear thermal expansion = 11.7x10-6 per oC.
Characteristic value of shrinkage is sh 0.0002
231
G2
The bridge carries a two-lane single carriageway road over a flood plain. The carriageway has 1m wide
shoulders and has a 2m wide footway on either side. A four-girder arrangement has been chosen and a
deck slab of 250mm has been assumed. The deck cantilevers 1.6m outside the centrelines of the outer
girders. The bridge is assumed to have 33m span with pinned end supports.
Figure G.1
G3
Single span multi-girder bridge structural arrangement
Design basis
The bridge is to be designed in accordance with the Bridge manual, AS 5100.6 and NZS 3101.
Clifton (2007a).
G3.1
Load combinations
Factors and load combinations of actions are given in the Bridge manual, tables 3.1 and 3.2 for SLS and
ultimate limit state (ULS) respectively.
G3.2
Factors on strength
232
G3.3

Structural steel:
grade 300
Concrete:
C40
Steel reinforcement:
grade 500
fy depends on the product material.

ASI design
320MPa for t 11mm; 300MPa for 11mm < t 17mm; and 280MPa for t
capacity tables
>17mm. For plates use:

320MPa for t 8mm; 310MPa for 8mm <t 12mm; 300MPa for 12mm <t 20mm; and 280MPa for t
>20mm.
Note that designers may wish to use grade 350 steel for better economy.
c'
For concrete, f = 40MPa

For steel reinforcement, fy = 500MPa
The modulus of elasticity for both structural steel and steel reinforcement is taken as Es = 200GPa
The modulus of elasticity of the concrete is given by NZS 3101 as:
E =
6900
c
1.5
'
3320 f +
r
2300
Refer to NZS 3101
clause 5.2.3
Ec = 32GPa
This 28-day value will be used for determination of all short-term effects and resistance and the modular
ratio is thus:
nS = 200 / 32 = 6.25
Refer to AS 5100.6,
clause 4.4.2
Ec
=
c,L
1+
32
= 10.50
1+ 2
cc
nL = 200 / 10.50 = 19
Surfaces in
The conservative value of cc is given in RRU 70 as 2.00
contact with the
G3.4
Surfaces with a
G3.4.1
Concrete

ground = 75mm
damp proof
membrane
between the
ground and the concrete = 50mm Elsewhere = 40mm

National Roads Board
(1984, section 2)
233
G3.4.2
Steel
For appropriate and cost-effective coating system for structural steelwork, the New Zealand steelwork
corrosion and coatings guide (El Sarraf and Clifton 2011), is used in conjunction with AS/NZS 2312 (SNZ
2002) and NZS 3404.1 (SNZ 2009).
G4
G4.1
Permanent actions
G4.1.1
G4.1.2
generally produces adverse effects and so it is based on nominal thickness + 55%. This follows
international practice; however, in New Zealand a minimum of 1.5kPa for surfacing plus 0.25kPa allowance
for services is used. Reference should be made to the Bridge manual, section 3.4.2.
DL 1.550.1323 4.63kPa
G4.1.3
G4.1.4
A nominal value of 2kN/m is assumed for each steel parapet.

Note that this value will be increased if a solid concrete crash barrier is used.
G4.2
Construction loads
For global analysis, a uniform construction load of 0.75kN/m2 is assumed during
Refer to section
casting and the weight of the temporary formwork is assumed to be 0.5kN/m2.
6.1.5 in the main
Additionally, wet concrete is assumed to have a density of 1kN/m greater than that of
part of the guide
hardened concrete; for a slab thickness of 250mm this adds 0.25kN/m2.

The total construction load is thus: CN 0.75 0.50 0.25 1.50kPa
G4.3
Traffic loads
G4.3.1
Road traffic
For the road carried by this bridge, the Bridge manual specifies the HN-HO-72
traffic loading but only HN72 has been considered here for simplicity.
G4.3.2
Pedestrian traffic
Pedestrian traffic is
represented by the
value in the Bridge
manual.
FP 5kPa
Refer to section 6.4.1 in the main part of the guide
This value can be reduced ti 4.0kPa when considered in conjunction with road traffic loading.
234
G4.3.3
Fatigue load
Clifton (2007a) is recommended for fatigue assessment.
G4.4
Thermal actions
G4.4.1
Refer to section
6.4.2 in the main
part of the guide
For a change of length in composite sections, the coefficient of linear thermal expansion is 11.7x10-6 per
C. This is then used for determining soil pressure on integral bridges or determining the expansion
length in the case of simply supported abutments.

G4.4.2
The vertical temperature difference given in the Bridge manual is used and the
Refer to Bridge
temperature difference will be considered to act simultaneously with the overall
manual, section
temperature change.
3.4.6
The effects of vertical temperature gradients shall be derived for both positive differential temperature
conditions (where the top surface is hotter than the average temperature of the superstructure) and
negative temperature differential conditions (where the top surface is colder than the average temperature
of the superstructure).
Note: The negative temperature variation is the same as that for bridge type 1 in figure 17.3 of AS 5100.2.
Figure G.2
G4.5
Seismic actions
Seismic actions have not been considered in this study. Refer to section 5 of the Bridge manual for
guidance on determining the horizontal force generated by seismic action on the bridge.
235
G5
Figure G.3
33.0m span girder

Top flange
600 x 25
Web
20
Bottom flange
800x 50
Top rebar
HD16-150
Bottom rebar
HD16-150
The overall girder depth is 1450mm. the cover to the centroid of the top and bottom reinforcement is
55mm. This is appropriate with reference to NZS 3101.
One alternative configuration to in-situ concrete slab is to include precast planks, which typically have prestressing in the bottom, and an in-situ slab with lighter reinforcement.
Note: The above section sizes are based on an initial few iterations to determine the section properties for
this worked example. For an actual bridge design the designer may wish to undertake further iterations in
order to optimise the cross-sections used and gain economy.
Figure G.4
The above bracing arrangements are assumed for this example.
236
G6
Beam cross section
G6.1
Section properties internal main girders
Figure G.5
Refer to section
7.1.1 in the main
part of the guide
Internal main girders effective width
For determination of stresses in the cross section and resistances of the cross section, the effective width
of the slab, allowing for shear lag is needed. The following calculations summarise the effective section
properties for the section considered.
The equivalent span for effective width is:
Refer to AS 5100.6,
Lc 33,000mm
b min
eff
Lc
;S;12t
clause 6.1.7
33,000
min
3000mm
s
;3700;12250
Refer to AS 5100.6,
clause 4.4.1
Properties for gross section in the mid-span are tabulated below.

Table G.1
Area
ENA height
Span girder
Unit
82,500
mm2
ENA
519
mm
2.81E+10
mm4
y, tf
30E+06
mm3
y, bf
54E+06
mm3
I
Z
Z
Section class
Not compact
Bending moment capacity(for +ve moment)
8400
kN.m
Slenderness check:
e,t_flange
=
e,t_flange
b t
f w
f
y,t_flange
250
2tf
600 - 20 280
=
=
12.28
2 25 250
Refer to AS 5100.6,
clause 5.1.2
237
e,web
=
d1
f
y,web
tw
250
1375 300
e,web =
=
250
75.3 20
e
= 14
y,flange
(1450 - 519 - 25)
re =
=
0.66
e
y,web
1375
60
60
re
Refer to AS 5100.6,
= 91
table 5.1
0.66
Plastic neutral axis = 154mm

(1450 -154 - 25)
rp =
= 0.92
1375
111
111
e
=
=
= 33.4
p,web 4.7r -1 4.7 0.92 -1
p
e
=8
p,flange
e,t_flange
12.28
14
y,flange
e,web
e
y,web
e,flange
75.3
91
>
= 0.88
= 0.83
e,web
e
y,flange
e =
e
= 12.28
flange
governs
y,web
flange
le le
p
S
Therefore, the bare steel section is not compact.

Table G.2
Composite cross section (short term) sagging (n =6.25)

S
Span girder
Unit
202,500
mm2
ENA
1145
mm
8.3E+10
mm4
Zy, c
150E+06
mm3
Zy, tf
272E+06
mm3
Zy, bf
73E+06
mm3
Area
ENA height
Section class
Not compact
Mp
24,671
kN.m
The cross section of the span girder is compact provided that the top flange is restrained by shear
connectors within the spacing limits given in AS 5100.6, clause 6.1.8. (the spacing of the shear stud is
more likely to be 150mm)
238
fy A
280 82,500
=
= 226mm
'
0.85fcb 0.85 40 3000
d ds
h
ds -d
250 - 226
-6
h
10
= 24,671kN.m
MP = Afy dg
=
82,500
280
1056
+
+
2
2
250
dg = 1450 +
- 519 = 1056mm
2
d =
h
Figure G.6
Section moment capacity compression zone entirely within the concrete slab
Note: During the construction of a composite bridge, it is quite likely that a beam will change its section
class, because the addition of the deck slab both prevents local buckling of the top flange and
significantly shifts the neutral axis of the section. Typically, a mid-span section could be compact after
casting the slab but not compact prior to this. As a consequence of this, checks at intermediate stages of
construction should be based on the relevant classification at the stage being checked.
Table G.3
Composite cross sections (long term) sagging (n =19)

L
Refer to section
Span girder
Unit
7.3.1 in the main
121,875
mm2
part of the guide
ENA
860
mm
Iy
5.8E+10
mm4
Section modulus, top slab
Zy, c
69E+06
mm3
Zy, tf
98E+06
mm3
Zy, bf
67E+06
mm3
Area
ENA height
G6.2
Effects of temperature difference and shrinkage
The primary effects of differential temperature through the depth of the cross section of a member are
considered. In addition, the secondary effects in continuous members, due to redistribution of the
moments and support reactions caused by the primary effects are also considered.
239
Longitudinal stresses and shear forces due to differential temperature effects are calculated by elastic theory
assuming full interaction between the concrete slab and the steel beam. The stiffness is based on the
transformed composite cross-section using a modular ratio appropriate to short-term loading and assuming
the concrete slab to have an effective width calculated in accordance with AS 5100.6, clause 4.4.1.
When the effects of shrinkage modified by creep adversely affect the structure, they are calculated in the
manner described for differential temperature effects, but using a modular ratio appropriate to long-term
loading. The beneficial effects of the creep of concrete are taken into account.
The effects of shrinkage in flexural design of girders are neglected as these effects are not adverse in the
case of simply supported bridges. However, the longitudinal forces due to the primary effects of shrinkage
and differential temperature are both considered in the design of the stud shear connector and transverse
reinforcement.
G6.2.1
For calculation of primary effects, the short-term modulus of concrete is used.

Ec = 32GPa
Es = 200GPa
For each element of the section, calculate stress as strain times modulus of elasticity, and then determine
force and centre of force for that area. The restraint moment in the inner beam, due to the characteristic
values of temperature difference is noted in the Bridge manual as:
o
T = (32 - 0.2d) C
o
T = 22 C
d = 50mm
The moment release stress is shown diagrammatically in figure G.7.

Figure G.7
The effects of the negative vertical temperature variation are insignificant and therefore have been
ignored.
G6.2.2
Shrinkage
The effects of shrinkage are calculated for the long-term situation.

The characteristic value of shrinkage is
sh
= 0.0002 and the long-term modular ratio is used.
240
Table G.4
Temperature and shrinkage induced restraint force

Force
Differential temperature axial fixity force
4313
kN
Shrinkage axial fixity force
1943
kN
G7
Global analysis
G7.1
Grillage model
Refer to AS 5100.6,
clauses 4.2 and 4.3
Refer to chapter 6 in the
A grillage model of the superstructure was created, with longitudinal elements
for composite girder and transverse elements for concrete slab.

As the bridge is simply supported at end supports, no concrete diaphragms were provided at both
abutments.
Grillage model of the superstructure
Support line
Dummy edge elements
(composite section)
Design line
3300mm
Transverse elements
(concrete slab)
Dummy edge elements
Support line
Figure G.8
370
0m
m
Note: the grillage model utilised a transverse elements comprising a rectangular slab 3300mm by 250mm.
Modern computer models automatically generate section properties; however, these should be checked
manually at least for torsional properties to suit the designers design assumptions.
G7.2
Construction stages
The whole of the deck span is concreted in one stage. The edge beams are concreted after this.
Separate analytical models are therefore provided for:
Stage 1: All steelwork, wet concrete (non-composite section properties)
Stage 2: Composite structure (long-term section properties), the weight of the edge beams is applied
Stage 3: Composite structure (short section term properties)
G7.3
Analysis results
The following results are for design values of actions for an internal beam, ie after application of
appropriate factors on characteristic values of actions.
241
For construction loading, results are given for the total effects at each of the construction stages. For
traffic loading the results are given for the combination of traffic and pedestrian loading for worst bending
effects at the location of the central span.
Note: The following results may not correlate with a slightly different model and hence should not be
viewed as absolute. They do, however, provide an indication of the order of magnitude expected.
G7.3.1
Stage 1
Selfweight of concrete
Construction load
Figure G.9
Stage 1 loading
Steel beam
selfweight
Concrete selfweight + Construction loading
ULS
Fz (kN)
M y (kN.m)
Fz (kN)
623
476
16.5
5867
4278
G7.3.2
SLS
My (kN.m)
Stage 2
Removal of construction loading
242
Figure G.10
Stage 2 loading
Parapet
selfweight
Footpath
construction
Carriageway surfacing
ULS
SLS
My (kN.m)
Fz (kN)
M y (kN.m)
Fz (kN)
278
206
16.5
2918
2162
G7.3.3
Stage 3
Table G.5
HN72 traffic loads for worst sagging at mid-span and worst shear at supports
ULS
SLS
My (kN.m)
Fz (kN)
M y (kN.m)
Fz (kN)
921
550
16.5
6052
3603
Note: HN-72 loading is assumed to be critical

Figure G.11
Stage 3 loading
Position of vehicle axle loads relative to the bridge abutments for the worst sagging moment at mid-span.
243
G8
Design values of effects are given below for certain design situations, for the design of the main girder
and cross girder beams. In practice, further situations for other parts of the structure would also need to
be considered.
G8.1
Effects of construction loads (ULS)
Generally, the effects of construction loads apply to different cross section properties, the cross sections
for the main girder beams are different at stages 1 and 2. The following tabulations summarise the forces
and moments at each stage and the stresses due to those effects, for selected cross sections. According to
the Bridge manual the load combination 5B is the most critical load case for construction stages.
Table G.6
Stress at mid-span
Bottom flange
M
Top flange
x, bf
x, tf
Top of slab
x, c
kN.m
kN
106mm3
MPa
106mm3
MPa
Stage 1
5867
623
54
109
30
-196
Stage 2
2918
278
67
44
98
-30
69
-42
-6
8785
901
Diff Temp - PE
155
106mm3
MPa
-
-232
-40
G8.2
Effects of traffic load plus construction loads (ULS)
Effects due to traffic actions are determined from the short-term composite section in mid-span.
Table G.7
Traffic load stress at mid-span

Bottom flange
M
Top flange
Top of slab
Fz
Zx, bf
Zx, tf
Zx, c
kN.m
106mm3
MPa
106mm3
MPa
106mm3
MPa
Construction
8785
901
HN72 traffic
6052
921
14,837
1822
155
73
83
238
-232
272
-22
-254
-40
150
-40
-80
The resulting stresses in the tables above indicate that the steel section remains elastic under construction
loading and full service load.
G9
Verification of bare steel girder during

construction
The paired beams are susceptible to lateral torsional buckling under the
Refer to figure 7.4 in the

weight of the wet concrete (ie before it hardens and provides restraint to the top flanges).
244
The beams are partially restrained against buckling by the bracing frames between each pairs at three
points in the span.
Use gross section properties for verification of buckling resistance.
G9.1
Evaluation of lateral restraint
If the length of the segment between braces satisfies the requirements in AS 5100.6, clause 5.3.2.4 the
nominal member moment capacity of a segment is taken as the nominal section moment capacity,
otherwise the member moment capacity is calculated using the expression in AS 5100.6, clause 5.6.1.
L
2rd A
f
250
2.5Zey
fy
80 + 50m
rz
if the segment is of unequal flanged I section
Refer to AS 5100.6,
clause 5.3.2.4
m = -1.0
Ls,max = 6500mm
fy = 280MPa
2.58E + 09
Iz
rz
=
=
A
= 177mm
82,500
6500
Ls,max
=
= 37
177
rz
80 +
2d A
f
50m
r=
Icz
25 2.5Zey
3
600
12
109
= Iz 2.58
d = 1412.5mm2
Af= 82,500m
m
Z
= min Z
;Z
ey
y,bf y,tf
rz
250
50m
= 37 80 +
= 30
2 0.17 1412.5 82,500 250

10
6
2.5 30
fy
= 20.6
280
= 0.17
= 30 10 6mm 3
2d A
f
250
2.5Zex
fy
= 20.6
M
M b s
=
G9.2
Verification
Refer to AS 5100.6,
clause 5.6.1
Maximum sagging bending moment at the mid-span under construction stage 1 load is 5867kN.m.
The flexural capacity of the main girder is given in AS 5100.6, clause 5.6 as follows:
Mb = msMs Ms
At the mid-span the bending moment assumed to be constant, therefore, m 1.00
Le = ktkLkrL
kt =
(LL)
1.0
kL =
1.0
245
kr = 1.0
Le = 5.00m
E = 200,000MPa
G 80, 000MPa
I = 2580
6
4
m
10 m
Iw 0
J = 31,952
3
4
m
10
m
Ms = 8400kN.m
2
2
EIy EIw
Mo
2 GJ + 2
Le
Le
=
2 200,000 2580
Mo
=
2
5000
6
10
80,000 31,952
3
10
10
-6
Mo = 22,819kN.m
Ms Ms
s = 0.6
+ 3-
Mo Mo
2
8400
8400
s = 0.6
+ 3= 0.84
22,819
22,819
M = 1.00 0.84 8400 = 7056kN.m

b
M = 0.9 7056 = 6350kN.m
b
M* = 5867kN.m < M = 6350kN.m - OK
b
G10 Verification of composite girder
Refer to figure 7.5 in the

G10.1 Sagging bending in main girder

the stages.
The bare steel section is not compact in bending and the composite cross section is compact. The effects
(stresses) in the cross section have been calculated on the basis of gross section properties for effects on
the steel beam plus effective section properties on the composite section.
The composite cross section is compact (PNA in the concrete slab) so the plastic resistance can be utilised.
The plastic bending resistance of the short-term composite section is 0.9 x 24,671=22,204kNm and the
total design value of bending effects is 14,837kNm. The cross section is satisfactory by inspection. It can
also be seen that the stresses calculated elastically, taking account of construction in stages, are
satisfactory, as shown in figure G.12.
246
Figure G.12
Bare steel and composite stresses in sagging
G11 Longitudinal shear
Refer to sections 6.1.7

and 7.7 in the main part
of the guide
shear connectors and the transverse reinforcement at the supports and at mid-span. (In practice,
intermediate values would also be verified, to optimise the provision of shear connectors.)
G11.1 Effects for maximum shear

Table G.8
ULS values at supports

Supports
623
kN
278
kN
921
kN
Table G.9
SLS values
Supports
476
kN
206
kN
550
kN
G11.2 Section properties

To determine shear flows, the parameter At yc It is needed for each section and stage.
Table G.10
Section properties
Web/top fl
Top fl/slab
A t y c It
A t y c It
m-1
m-1
Steel section
0.490
Long term section
0.636
0.487
Short term section
0.675
0.626
247
G11.3 Shear flow at ULS

Force at web/top flange junction
At abutment
623 x 0.490 + 278 x 0.636 + 921 x 0.675 =
1104
kN/m
278 x 0.487 + 921 x 0.626 =
712
kN/m
206 x 0.487 + 505 x 0.626 =
416
kN/m
At abutment
G11.4 Shear flow at SLS

At abutment
G11.5 Web/flange weld

For 6mm throat SP fillet weld t =6mm
w
6mm
0.835kN / mm
Resistance of two welds = 1670kN/m > 1104kN/m, shear flow at top flange OK
Refer to section 7.7.3 in
G11.6 Shear connectors
the main part of the guide
Shear stud connectors 19mm diameter, 150mm long are assumed, with f =410MPa.
u
2
fuc
f0. =
63d
ks
bs
or
2
'
f0. =
f E
63d
ks
bs cy c
'
fcy
= 40MPa
Refer to AS 5100.6,
clause 6.6.4.4
d =
16mm bs
fcu =
410MPa
'
Ec = 5050 fcy = 5050 40 = 32,000MPa
Therefore the design resistance of a single headed shear connector is:

f = min 0.63
ks
2
410;0.63
19
19
40 32,000= 93.25kN
Longitudinal shear forces due to the primary effects of shrinkage or differential temperature are assumed
to be transmitted across the interface between the steel beam and the concrete slab by shear connectors
at each end of the beam, ignoring the effects of bond.
In the absence of a more accurate analysis, the forces on the connectors may be calculated by assuming
that the rate of transfer of longitudinal force varies linearly from a maximum at the end of the beam to
zero at a distance from the end equal to the total effective width of the slab. Alternatively, where stud
shear connectors are used, the rate of transfer of force may be assumed to be constant over a distance
from each end of the beam equal to one fifth of the span of the beam.
In this exercise, it is assumed that longitudinal shear flow is constant over the span of the beam.
248
L* Ls
Ls 0.55nf
ks
1.0
563kN / m
1943 5
* 416kN / m
L
332
kN / m
Shrinkage PE
Refer to AS 5100.6,
clause 6.6.3.2
then a row of 2 studs has a design resistance of:
0.5593.253

769kN/ m 563kN/ mOK
Ls
L
0.20
Refer to section
7.7.7 in the main
part of the guide
G11.7 Transverse reinforcement

Consider the transverse reinforcement required to transfer the full shear resistance of the studs, ie
522kN/m as well as the maximum ULS shear flow, ie 1154kN/m.
Figure G.13
Transverse reinforcement critical shear planes

Refer to AS 5100.6,
clause 6.6.5.2
AS* 5100.6, clause 6.6.5.2 as follows:
Lp 0.9us 0.7Ats fry

and
*
'
Lp 0.15ufc
assumed HD16-150mm:
s 1MPa
u 2150 300 600mm
Ast 2201 150 2680mm2 / m
fr 450MPa
y
0.72680450
L min 1.0 0.96001.0
;1.0 0.15600401384kN / m
3
10

*
1104kN / m
1.3519435
1303kN / m
332
Shrinkage PE
Lp
* 1303kN / m 1384kN / mOK

Lp
249
G12 Fatigue assessment

G12.1 Basic loadings for fatigue design
Refer to sections 6.4.2

and 7.11.1 in the main
part of the guide
There are two basic loads given, as clause 6.9 of AS 5100.2, being:
The modified individual A160 heavy axle load from AS 5100.2, as shown in below
Figure G.14
Modified individual A160 heavy axle loading
The modified individual M1600 moving traffic load from AS 5100.2, without the UDL component, is shown
in figure G.15.
Figure G.15
Modified individual M1600 moving traffic loading
The number of fatigue stress cycles used for the calculation of the fatigue capacity of the structural
1
(Current number of heavy vehicles per lane per day) x 4 x 104 x (route factor).
(Current number of heavy vehicles per lane per day) x 2 x 104 (L-0.5) x (route factor).
Dynamic load allowance is given as following:

250
For positive bending moments, L is the actual span in which the bending moment is being considered.
For negative moment over interior supports, L is the average of the adjacent spans.
For end shear, L is the actual span.
For reactions, L is the sum of the adjacent spans.
For cross-girders, L is twice the longitudinal spacing of the cross-girders.
G12.2 Range of effects due to passage of fatigue vehicle

By inspection it is clear that the fatigue loading of M1600 is more critical, therefore the fatigue loading is:
0.70 (M1600 without UDL) (1 + )

And corresponding total number of cycles is:
n = NHVDper lanex 2104 L0.5 route factor

NHVD 1500
Table G.11
route factor 0.3

Span
M (kN.m)
y
Range
2236
Table G.12
Worst shear effects

Span
Fz (kN)
Range
336
G12.3 Assessment of structural steel details

G12.3.1 Design stress range mid-span
For the mid-span the length of the effective span is L 33.0m .
4
n = 1500 2 10
33
-0.5
10
6
0.3 = 1.57
At mid-span there is negligible stress range in the top flange. The stress range in the bottom flange is
f* =
2236
73
= 31MPa .
As f* 31MPa 27MPa fatigue assessment is required.

The worst detail category that might apply is for the weld between the bottom flange and the web, which,
for manual continuous weld, is category 100 (AS 5100.6, table 13.5.1-B).
Design value of fatigue strength ffc is:
n 1.57106 5106
1.0
251
frnc 100
210
f
2 13
= 100
f
rnc
fc
1
6
3
f f OK
108MPa
fc
1.57
G12.4 Assessment of shear connection

The design value of the stress range in shear studs is 425MPa times the ratio of the
Refer to AS 5100.6,
clause 13.7.2
longitudinal shear load on the stud to the nominal static strength specified in AS 5100.6, clause 6.6.3.
G12.4.1 Shear at supports
The range of vertical shear force at the support is 336kN.
At the pier, the studs are 19mm diameter, in rows of two at 200mm spacing.
L* 3360.626 210kN/ m
At yc
It
0.626m1
Ls 513kN / m
1.0
Therefore,
2100.2
*
*
L
f 425
2
425 96MPa
s
513
Ls
The fatigue strength of the shear stud is given in AS 5100.6, clause 13.6.3 as following:
1
1
2.081022 8
2.081022 8
104MPa
f
6
n
1.5710
s
f f OK
Note: that designers may select the transverse reinforcement spacing to avoid clashed with shear studs.
252

Steel-Concrete Composite Bridge Design Guide September 2013

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Steel-Concrete Composite Bridge Design Guide September 2013

Uploaded by

Copyright:

Available Formats

Steel-concrete composite bridge design

Raed El Sarraf, HERA, Auckland, New Zealand

NZ Transport Agency research report 525

ISBN 978-0-478- 40769-3 (electronic)

Keywords: bridges, composite, concrete, construction, cost-effective, design, durability, efficiency,

studs, steel, Steel Bridge Development Group, sustainability, vibration, welding.

An important note for the reader

Shane Culham of Culham Engineering

Jacques De Reuck of Metal Spray Suppliers (NZ) Ltd

Bob Delacey of George Grant Engineering

Alistair Fussell of Steel Construction New Zealand

Phil Gaby of Holmes Consulting Group

Kenyon Graham of SKM

Matthew Holland of Tenix Robt Stone

Donald Kirkcaldie of Opus

Peter Lipscombe of URS

Jamie Macredie of Steltech

Willie Mandeno of Opus

Bruce Melsop of Eastbridge

Amin Momtahan of AECOM

David Moore of Grayson Engineering

Will Pank of BECA

Rob Presland of Holmes Consulting Group

Anand Reddy of Holmes Consulting Group

Craig Ross of Napier Sandblasting

Dr Wolfgang Scholz of HERA

Mike Sullivan of D&H

Tony Vos of Altex

Peter Wiles of Opus

Abbreviations and acronyms

British Standards Institution

computer-aided engineering tools

European Committee for Standardization

fracture critical members

inorganic zinc silicate

New Zealand Transport Agency

Road Research Unit, National Roads Board (predecessor of Transit NZ)

serviceability limit state

Standards New Zealand

ultimate limit state

Typical composite bridge configurations ........................................................................... 10

Multi-girder bridges .................................................................................................. 10

Ladder deck bridges .................................................................................................. 14

Deck slab ................................................................................................................... 20

Shear connection ...................................................................................................... 21

Dealing with curvature and skew ............................................................................... 21

Benefits of steel-concrete composite construction ............................................................. 30

Why specify a composite bridge? ............................................................................... 30

Sustainability benefits ............................................................................................... 30

Economic benefits ..................................................................................................... 32

New Zealand case studies .......................................................................................... 34

Design concept ......................................................................................................... 37

Preliminary design .......................................................................................................... 38

Design for construction ............................................................................................. 38

Design for in-service maintenance ............................................................................. 42

Choice of structural configuration .............................................................................. 44

Preliminary sizing material selection ........................................................................ 45

Preliminary sizing multi-girder bridges ................................................................... 46

Preliminary sizing ladder decks ............................................................................... 48

Design standards ............................................................................................................ 49

The Bridge manual .................................................................................................... 49

NZS 3404 .................................................................................................................. 50

NZS 3101 .................................................................................................................. 51